|UFDC Home||myUFDC Home | Help|
This item has the following downloads:
1 PREDICTORS OF PROPER NAME RETRIEVAL DEFICITS IN OLDER AGE By MEAGAN T. FARRELL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 20 12
2 20 12 Meagan T. Farrell
3 To my parents, for their unwavering love and encouragement, and to my advisor, Lise Abrams, for her invaluable insight throughout this process
4 ACKNOWLEDG MENTS I would like to thank my committee for their thoughtful and useful feedback on my dissertation proposal, which undoubtedly strengthened the methodology and interpretation of the project. I also thank Dr. Chris Votolato, director of behavioral health, and other helpful personnel at Shell Point Village retirement community in Fort Myers Florida for their assistance in the recruitment of older adult participants. It was a pleasure using their facilities during data collection, overlooking pristine SW Flo rida environs while eliciting TOT s
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 11 Aging and Lexical Access ................................ ................................ ....................... 12 Theories of Proper Name Specificity ................................ ................................ ...... 15 Aging and Proper Name Retrieval ................................ ................................ .......... 26 Priming and Tip of the Tongue (TOT) States ................................ ......................... 33 The Current Research ................................ ................................ ............................ 40 2 PILOT STUDY ................................ ................................ ................................ ........ 43 Method ................................ ................................ ................................ .................... 43 Data Collection and Participants ................................ ................................ ...... 43 Materials ................................ ................................ ................................ ........... 43 Procedure ................................ ................................ ................................ ......... 45 Results ................................ ................................ ................................ .................... 48 3 EXPERIMENT 1: PHONOLOGICAL INFLUENCES ON NON NAME AND PROPER NAME TOTS ................................ ................................ ........................... 54 Specific Aims ................................ ................................ ................................ .......... 54 Specific Aim 1 ................................ ................................ ................................ ... 54 Specifi c Aim 2 ................................ ................................ ................................ ... 55 Specific Aim 3 ................................ ................................ ................................ ... 56 Method ................................ ................................ ................................ .................... 57 Participants ................................ ................................ ................................ ....... 57 Materials ................................ ................................ ................................ ........... 58 Procedure ................................ ................................ ................................ ......... 61 Results ................................ ................................ ................................ .................... 64 Initial Responses ................................ ................................ .............................. 65 TOT Incidence ................................ ................................ ................................ .. 66 Main analyses ................................ ................................ ............................ 66 Other proportional measures of TOT incidence ................................ ......... 71 Correct response rates ................................ ................................ .............. 72 TOT Resolution ................................ ................................ ................................ 75
6 Phonological priming effects on TOT resolution ................................ ......... 76 Effects of syllable frequency on primed TOT resolution ............................. 77 Partial name retrievals ................................ ................................ ............... 78 Discussion ................................ ................................ ................................ .............. 79 4 EXPERIMENT 2: SEMANTIC INFLUENCES ON NON NAME AND PROPER NAME TOTS ................................ ................................ ................................ ........... 95 Specific Aims ................................ ................................ ................................ .......... 96 Specific Aim 4 ................................ ................................ ................................ ... 96 Specific Aim 5 ................................ ................................ ................................ ... 97 Method ................................ ................................ ................................ .................... 98 Participants ................................ ................................ ................................ ....... 98 Materials ................................ ................................ ................................ ........... 99 Procedure ................................ ................................ ................................ ....... 100 Results ................................ ................................ ................................ .................. 101 Initial Responses ................................ ................................ ............................ 101 Sem antic Similarity Ratings ................................ ................................ ............ 102 TOT Incidence ................................ ................................ ................................ 103 Main analyses ................................ ................................ .......................... 103 Co rrect response rates ................................ ................................ ............ 105 TOT Resolution ................................ ................................ .............................. 109 Semantic priming effects on TOT resolution ................................ ............ 110 Effects of syllable frequency on TOT resolution ................................ ....... 110 Discussion ................................ ................................ ................................ ............ 111 5 GENERAL DISCUSSION ................................ ................................ ..................... 123 APPENDIX A NON NAME TARGET, PRIME AND UNRELATED QUESTIONS ........................ 140 B PROPER NAME TARGET, PRIME, AND UNRELATED QUESTIONS ................ 161 LIST OF REFERENCES ................................ ................................ ............................. 182 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 189
7 LIST OF TABLES Table page 2 1 Means and standard deviations for familiarity ratings and semantic similarity ratings in the pilot study ................................ ................................ ...................... 51 3 1 Descriptive characteristics for youn g and older participants in Experiment 1 ..... 84 3 2 Sample non name targets with phonological, semantic, and unrelated primes .. 85 3 3 S ample proper name targets with phonological, semantic, and unrelated primes ................................ ................................ ................................ ................. 85 3 4 Mean word length of each type of target and prime question ............................. 86 3 5 Percentage of each response type as a function of age group and target type. ................................ ................................ ................................ .................... 86 3 6 type, prime cond ition, and first syllable frequency. ................................ ............. 87 3 7 function of target type and prime condition. ................................ ........................ 88 3 8 type and prime condition. ................................ ................................ ................... 89 4 1 Descriptive characteristics for young and older part icipants in Experiment 2 ... 115 4 2 Percentage of each response type as a function of age group and target type. ................................ ................................ ................................ .................. 116 4 3 Young and ol type, prime condition, and first syllable frequency. ................................ ........... 117 4 4 target type, prime condition, and first syllable frequency. ................................ ........... 118 4 5 type and prime condition. ................................ ................................ ................. 119
8 LIST OF FIGURES Figure page 1 1 Representation of Proper Names and Non Names according to NST ............... 42 2 1 Familiarity rating scale for non name targets ................................ ...................... 52 2 2 Fam iliarity rating scale for proper name targets ................................ .................. 52 2 3 Semantic similarity rating scale for non name targets ................................ ........ 53 2 4 Semantic similar ity rating scale for proper name targets ................................ .... 53 3 1 Trial structure for Experiment 1 ................................ ................................ .......... 90 3 2 Age differences in TOT incidence as a function of target type and first syllable frequency. ................................ ................................ .............................. 91 3 3 Age differences in phonological priming of TOT incidence. ................................ 92 3 4 Age differences in correct known responses as a function of target type and first syllable frequency. ................................ ................................ ....................... 93 3 5 Effects of syllable frequency on primed TOT resolution of non name and proper name targets (collapsed across age group). ................................ ........... 94 4 1 Trial structure for Experiment 2 ................................ ................................ ........ 120 4 2 Age differences in TOT incide nce as a function of target type and first syllable frequency. ................................ ................................ ............................ 121 4 3 Age differences in correct response rates as a function of prime condition, target type, and syllable frequency. ................................ ................................ .. 122
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PREDICTORS OF PROPER N AME RETRIEVAL DEFICITS IN OLDER AGE By Meagan T. Farrell December 20 12 Chair: Lise Abrams Major: Psychology word finding failures, or tip of the tongue (TOT) states. Rela tive to other words, proper names seem to elicit more frequent and socially consequent TOTs, particularly for older adults. The current dissertation aimed to disambiguate phonological and semantic contributions to the disproportionate impairment for proper name retrieval in older age. Young and older adults read general knowledge questions whose target answers were famous names (e.g., Alfred Hitchcock ) or low frequency noun non names (e.g., arson ). Half of the targets began with a high frequency (HF) first syllable and half began with a low frequency (LF) first syllable. In Experiment 1, target questions were preceded or target (e.g., Albert Einstein arbitrary ), or an unre lated question. In Experiment 2, prime questions contained a name or word that was semantically related to the target (e.g., Steven Spielberg pyromania ). Results confirmed a disproportionate age deficit for names: older adults had more TOTs than young adu lts, and this age difference was more pronounced for proper names than non names. However, the age related impairment for names was moderated by first syllable frequency in that exaggerated
10 age deficits only occurred for names with HF first syllables. Furt her, older adults (but not young) showed opposing influences of syllable frequency for non names and names, experiencing more TOTs for non names with LF first syllables but fewer TOTs for proper names with LF first syllables. With respect to priming, phono logical primes reduced TOTs, increased correct retrievals, and boosted TOT resolution, but only for older adults and for non names. Conversely, semantic primes did not influence TOTs names. T hese findings suggest that age differences in TOTs for non names and proper names are driven by different characteristics. For non names, TOTs represent failed activation of LF phonology, whereas for proper names, TOTs result from increased complexity in the structure of names' lexical representations where first and last names are shared by other people. T hese results provide clarity and specificity to models of proper name retrieval and cognitive aging.
11 CHAPTER 1 INTRODUCTION The act of retrieving a fa to which most people give little notice until they are unable to recall the name of someone they know. These frustrating and embarrassing lapses in memory seem to occur more frequently in late life (e.g., Burke, Locantore, Austin, & Chae, 2004; Cohen & Faulkner, 1986; Cross & Burke, 2004; Evrard, 2002; Fogler & James, 2007; but see James, 2006; Maylor, 1997), with name forgetfulness easily emerging as one of the most prevalent cognitive complaints among the elderly (e.g., Cohen & Burke, 1993; Cohen & Faulkner, 1986; Evrard, 2002; Maylor, 1997). Perhaps due to the social salience of names, cognitive psychologists and neuropsychologists have shown an increasing interest in understanding the mechanisms that und erlie the production of proper names and the conditions that promote retrieval failures. Theoretically, the primary question of interest is how the representation of proper names differs from other word classes at the neural and/or cognitive level, and how this divergence influences the ease with which proper names are retrieved compared to other types of words. Extant neuropsychological and behavioral evidence suggests that proper names induce more retrieval failures (e.g., Bredart & Valentine, 1998; MacKa y & Burke, 1991; Semenza, 2006; Semenza, 2009), are named more slowly (e.g., Evrard, 2002), and are more difficult to learn (e.g., Brennen, 1993; Cohen & Burke, 1993; Cohen & Faulkner, 1986) than other words (but see Hanley, 2011), although the source of t his selective impairment remains unresolved. As such, the global goals of the present research were to: 1) investigate how the representation and retrieval of proper names differ from other categories of words, 2) elucidate why this dissociation makes prop er names more difficult to retrieve
12 and produce, and 3) address why the disproportionate difficulty with names seems to be exacerbated by aging. More specifically, the project examined both phonological and semantic influences on proper name and non name 1 retrieval among young and older adults. Aging and Lexical Access Regardless of word class, the process of retrieving and producing a single word is deceptively complex, providing a number of opportunities for production processes to go wrong. Most cogniti ve models of speech production assume that semantic (meaning based), lexical (syntax based), and phonological (sound based) representations must be activated during the produc tion of a single word (e.g., Bock & Griffin, 2000; Dell, 1986; Garrett, 1988; Lev elt, Roelofs, & Meyer, 1999; MacKay, 1987; Rapp & Goldrick, 2000). For example, production of the word dog requires the activation of its semantic and/or perceptual features (e.g., an animal, covered in fur, companion), its lexical representation or lemma (e.g., the specific word that best embodies the activated semantic features), and its phonological features (e.g., the individual phonemes within the word). Distinct representations are accessed at each level in the language system, and speech errors resul t when a speaker fails to activate the correct representations in the correct order. For example, semantic substitution errors occur when top down activation from the semantic level results in the erroneous selection of a lemma that is semantically similar to an intended target word (e.g., nun for priest ). Sound based errors or slips of the tongue occur when incorrect syllables or phonemes get activated in the 1 Nouns that are not neuropsychological and psycholinguistic literatures. However, because the nouns selected for this experiment are especially rare, low frequency nouns, the term common name seemed like an typ e of word that is not defined as a proper noun.
13 phonological system (e.g., dock the loor for lock the door ) Finally, word retrieval failures occu r when a speaker is unable to access the semantic, lexical, or phonological representation of a known word, thereby obstructing production of an intended target word. Tip of the tongue (TOT) states represent a unique situation where a speaker is unable to MacKay, Worthley, & Wade, 1991; Levelt, 1989; but see Caramazza & Miozzo, 1997 for an alternative view where TOTs result from incomplete access of the lexical representation ). T he frustration that accompanies TOT states is thought to reflect successful selection of a specific lemma, giving the speaker the sensation that the word is on the verge of being produced. Although all language users experience the occasional TOT, olde r adults appear to be more susceptible to these temporary memory lapses, as demonstrated by both diary studies of naturally occurring TOTs (e.g. Burke et al., 1991; Heine, Ober, & Shenaut, 1999) and laboratory studies comparing TOT incidence rates among yo ung and older adults (e.g., Heine et al., 1999; Vitevitch & an age related reduction in the transmission of activation between various levels of a explanation referred to as the Transmission Deficit Hypothesis (TDH; e.g., MacKay & Burke, 1990). According to TDH, connections weakened with age, thereby reducing the transmiss ion of activation between all nodes. Linkage strength may become so weak that the amount of activation transferred throughout nodes in the system may be inadequate to fully activate a targeted node.
14 The assumptions made by TDH are consistent with neurobio logical characteristics of aging in that older adults experience white matter atrophy and truncated myelinated fibers, which in turn constrain neural connectivity (e.g., Burke & Shafto, 2008; Hedden & Gabrieli, 2004). Although declines in connection integr ity are diffusely distributed within TDH, the organizational hierarchy of the nodes renders some links more vulnerable to transmission deficits than others. The retrieval of a given phonological node from the phonological network is particularly susceptibl e to transmission failure because it is its phonology. In contrast, the semantic system is more resistant to age related transmission deficits because it is characterize d by multiple converging and redundant connections among semantic and lexical nodes. Empirically, TDH's reduction in the transmission of activation in old age also offers a language specific explanation for why processes that require speeded phonological r etrieval, such as picture naming, are more susceptible to age related declines than processes relying solely on semantic information such as vocabulary (e.g., Mortensen, Meyer, & Humphreys, 2006). Given the hierarchical architecture of the language system, it seems plausible that certain characteristics of words and their phonological forms may exacerbate age related retrieval deficits for healthy older adults or those with pathological cognitive deficits (e.g., Balota & Ferraro, 1996; Vitevitch & Sommers, 2003). Indeed, relevant evidence suggests that age differences in language processes may be linked to lexical (structural influences of the whole word) as well as sublexical (structural influences of phonological components contained within a word, e.g., p honemes and syllables) properties of words (e.g., Carreiras, Baquero, & Rodriguez, 2008; Farrell & Abrams,
15 2011; Spieler & Balota, 2000; Vitevitch & Sommers, 2003). Because of age related changes in phonological level processes, a few studies have reported differential effects language) and neighborhood density (the number of words in the lexicon that are phonologically similar to a target word) as a function of age ( e.g., Carreiras et al., 2008; Farrell & Abrams, 2011; Vitevitch & Sommers, 2003; but see Spieler & Balota, 2000). In examining the incidence of TOTs for non names, Farrell and Abrams (2011) found that older adults, but not young adults, had more TOTs for w ords beginning with low frequency (LF) first syllables (e.g., ominous ) relative to words with HF first syllables, (e.g., decanter ). Importantly, age differences in TOT incidence only occurred when participants attempted to retrieve words with LF onset syll ables, suggesting that older phonological components of a word. Assuming that older adults have greater difficulty with words and sounds used infrequently and/or are less conn ected with other representations in the language system, they might show a particular susceptibility to word types that embody both of these characteristics, i.e., proper names. Theories of Proper Name Specificity Some of the earliest research on proper n ame retrieval stemmed from an attempt biographical information about them. To account for this asymmetry, Bruce and Young (1986) proposed a model of face recognition/pe rson identification that has been used to guide a number of empirical investigations into proper name retrieval. According to this perspective, the retrieval of a name is the last step in a sequential series of events required to recognize and name a known person. Upon encountering a face, a person
16 must activate (in order) long term perceptual codes for a familiar face (Face Recognition Units; FRUs), then activate Person Identification Units (PINs) which contain biographical and semantic features about the person (age, occupation, nationality), and finally activate the name nodes corresponding to the identified person. Thus, name retrieval relies on prior activation of semantics and is therefore more susceptible to error than the retrieval of biographical fa cts (see Burton, Bruce, & Johnston, 1990, for later adaptation that allows for semantic activation and name retrieval to occur in parallel). Indeed, a number of studies have shown that names are harder to access than other types of information about a per son (e.g., Johnston & Bruce, 1990; Young, Ellis, & Flude, 1988; Young, McWeeny, Ellis, & Hay, 1986). For example, when asked to make judgments about two pictures of people, participants are faster when the task requires them to compare semantic features of the two targets (e.g., do they have the same occupation, are they both alive?) compared to when they are asked to confirm whether they share the same name (e.g., Johnston & Bruce, 1990; Young et al., 1988) Similarly, a number of studies have shown that n ames are more difficult to learn than other features about a person (e.g., Cohen, 1990; Cohen & Faulkner, 1986). This effect persists even when the same phonological word form is used; it is easier to remember that a person works as a baker than to remembe r his name is Mr. Baker (e.g., McWeeny, Young, Hay, & Ellis, 1987), an effect referred to as the Baker baker paradox (e.g., Cohen & Burke, 1993; Semenza, 2009). All of this evidence is taken to suggest that names are stored separately from other types of p erson specific semantic information and that the latter must be successfully activated prior to production of a name.
17 to more global speech production processes in that the production of names seems to require additional effort relative to other word types. A number of studies have shown that proper names are associated with more TOTs than other classes of words (e.g., Burke et al., 1991; Bredart & Valentine, 1998; Evrard, 20 02; MacKay & Burke, 1991; but see Hanley, 2011), and various characteristics of proper names have been proposed to account for this selective deficit. Most of the proposed characteristics relate to the mes, which presumably create a separation between names and other words within the mental lexicon. By nature, names may be more arbitrary (e.g., Semenza, 2006), more specific (e.g., Bredart, Valentine, Calder, & Gassi, 1995; Semenza, 2006; Semenza, 2009), less descriptive (Fogler & James, 2007), less meaningful (e.g., Cohen & Burke, 1986), and less frequent (Cohen & Burke, 1993; Conley, Burgess, & Hage, 1999), all of which may contribute to the isolation of names in the memory system, therefore making them more difficult to access Evidence for at least partially distinct neurocognitive systems dedicated to proper name and non name word retrieval comes from neuropsychological case studies showing selective impairment, and in rarer cases, selective sparing of proper names among aphasics (e.g., Semenza, 2006; 2009). In the most common type of proper name anomia, patients have an inability or profound difficulty when producing the names of people but not when asked to name objects or other types of words. Less commonly, neurological damage can cause the reverse pattern, where a patient is relatively preserved during proper name retrieval but has marked difficulty during the
18 retrieval of other word types. This double dissociation is taken to suggest that there m ust be some separation between the brain regions dedicated to the processing of proper names and other nouns. The infrequency of these cases, however, indicates that proper name and non name retrieval mostly rely on common neural substrates and only rare and highly localized lesions selectively target a given word class. Working off existing cognitive and psycholinguistic theories (e.g., Cohen & Burke, 1993), Semenza argues that independent processing pathways are engaged during the retrieval of names vers us non because of an inherent weakness in the relationship between a proper name (the phonological label) and its reference (the person it corresponds to) Unlike other types of words, a proper n ame does not designate meaning and refers instead to a single entity, an entirely unique instance. In contrast, the lexical representation of a non name refers to an interrelated web of attributes that signify and define the reference. In other words, prop er names refer to individuals whereas other words refer to categories; one may encounter 500 different pencils but each time call it a pencil, whereas a unique identifier must be stored for each new person. The distinction between the representations of pr oper names and other words may be theoretically comparable to the dissociation between episodic and semantic memory in that the former refers to the storage of facts about individuals and events, and the latter refers to the storage of shared knowledge tha t is reused in various different contexts (see also Cohen & Faulkner, 1986). Overall, there seems to be evidence for separate processing streams dedicated to proper name and non name retrieval; further, some inherent mechanism makes the proper name stream more resource demanding than the other.
19 Similar theories about the semantic features of proper names have been used to explain proper name retrieval difficulties among non aphasic speakers. For example, previous research suggests that names are more diffi cult to learn (and therefore remember) than other types of information because the phonology of a name does not symbolize semantic information. When a new person is encountered, biographical information about the person is encoded more strongly than the pe existing semantic network (e.g., the courtroom, law school, personality characteristics, etc.). The role of semantic meaning (or lack of semantic meani ng) in the disproportionate difficulty with proper name memory is supported by a study reviewed by Cohen and Burke (1993), where the experimenters were able to reverse the s were used as a memory cues, meaningful information about a person (e.g., is a potter) was more easily retrieved than meaningless names (e.g., Mrs. Ryman), consistent with previous research However, the opposite effect occurred when the meaningless label s were swapped, so that a name with meaning (e.g., Mrs. Potter) was more easily retrieved than a meaningless occupation (e.g., is a ryman). Thus, the learning of new names may be hindered by the fact that names characteristically lack meaning, and as such, they form less stable connections with other words in the lexicon and are more difficult to access on subsequent retrieval attempts. Although there may be considerable differences between the semantic representations of proper names compared to other wor ds, recent research has suggested that proper names may form a tight person specific network within semantic
20 memory (e.g., Burgess & Conley, 1998; Conley et al., 1999). The hyperspace analogue to language (HAL) is a computational model of semantic memory u sed to conceptualize the distribution of concepts in semantic space. Using the HAL model, Burgess and Conley (1998) found that proper names draw away from other word classes and form a separate but tight semantic cluster. Relative to non names, the semanti c neighborhood density of proper names is higher, suggesting that the semantic representations of names may form closer connections to each other than do words of other classes. In support of this idea, Conley et al. (1999) found that HAL neighborhood dens ity was a strong predictor of word naming latencies when participants were asked to pronounce first names aloud, suggesting that semantic features also influence successful production of names. Beyond the semantic level, the greater difficulty with proper name retrieval may be linked to a lexical level feature, frequency of use (e.g., Cohen & Burke, 1993). The frequency effect is one of the most well established findings in the speech production literature, where low frequency words are produced more slowl y and less accurately than high frequency words (e.g., Besner, Moroz & O'Malley, 2011; Jescheniak & Levelt, 1994; Oldfield & Wingfeild, 1965; Stemberger, 1984) and are also more likely to cause a TOT state (e.g., Burke et al., 1991). Proper names possess an overall lower frequency than other types of words (e.g., Fogler & James, 2007), and even the most common surnames (e.g., Smith ) are probably used and encountered less frequently than most other words, including biographical information about the person named Smith (e.g., lives in an apartment, is a teacher) Conley et al. (1999) compiled two corpora (one based on phone book entries, the other based on internet text usage) to
21 measure the frequency of first names and its influence on naming times. Name fre quency emerged as a significant predictor of pronunciation latencies. However, interpretations from this study are limited by the fact that frequency measures were based on the frequency of first names among all first names, not on the frequency of names a mong all words in the lexicon. The latter measure would be a more accurate and ecologically valid indicator of the frequency with which names are produced during everyday speech and would therefore have greater explanatory value in accounting for the diffe rence between proper names and non names. Additionally, differences between proper nouns and other noun types persist when the exact same phonological forms are retrieved or learned (e.g., McWeeny et al., 1987) suggesting that that frequency of use is not the only reason for the increased difficulty with proper names. The difficulty with proper name retrieval can also be examined within a general framework of language processing that represents linguistic units as nodes within a complex connectionist netw ork ( e.g., MacKay & Burke, 1991). According to Node Structure Theory to the relatively weak connection formed between the visual image of a person and his/her name that occurs during l earning. Figure 1 depicts the different networks associated with learning the name of a person ( John Baker ) and learning his occupation ( baker ) visual concept node representing t he person and noun phrase node that ( John Baker Figure 1 1 ). Solitary connections also form between the proper noun phrase node and the two lexical nodes for the first ( John ) and last ( Baker ) name. In contrast, once biographical information about that person is
22 learned (i.e., John is a baker ), a connection is formed between the visual concept node for John (face) and the lexical node representing the occupation ( baker, Figure 1 1 ). The noun node for baker (in th e semantic system) forms connections with an existing network of information about bakers (i.e., makes bread, wears a hat, etc. ). As a result, this entire network of nodes also becomes connected to the visual concept node for John. Therefore, the difficult y that arises from attempted retrieval of the name could result from the solitary connection between the visual image of a person to the proper noun phrase node and to the corresponding lexical nodes for the first and last name. Conversely, multiple direct and indirect connections allow for more stable links between the visual concept (face node) and corresponding biographical information about the individual. The discrepancy in connection strength can also account for cueing asymmetry: When we recall or ar e provided biographical information about an individual about that person. According to NST, this asymmetry is due to the convergence of activation on nodes that represent biographical information versus the isolated lexical level representation of names (see also Shafto & MacKay, 2000, for expansion of theory to other name phenomena). In sum the retrieval of proper names might be hindered by multiple isolated links within the lexical system. This effect might be exacerbated if proper names also contain rare and infrequently accessed phonological units, which have weaker links to the lexic al/semantic level (discussed in greater detail below with regards to aging).
23 However, evidence for a disproportionate difficulty with names is not unequivocal. A primary criticism of extant research is that much of the data showing higher TOT rates for na mes versus non names come from diary studies (Burke et al., 1991; Cohen & Faulkner, 1986), which rely on the self reports of participants and are therefore susceptible to memory biases. TOTs for non names may be underrepresented in diary studies because pe ople can often supply an alternative word with a similar meaning, making the TOT experience less salient and therefore less memorable than a TOT for be offset by calling it a carafe or fancy jug, thereby masking the temporary lapse in memory. Similar options are not available when trying to retrieve the name of specific have argued that la boratory TOT studies underestimate the rate of TOTs for proper names because they cannot use acquaintance names, which are the most common source of retrieval failures. Only a handful of studies have strategically compared TOTs for names and non names in a n experimentally controlled design (e.g., Burke et al., 1991; Evrard, 2002; Hanley, 2011), and variability in the methodologies and dependent measures have hindered comparability across studies and challenged strong empirical conclusions. For example, Evra rd (2002) elicited TOTs by asking participants to name pictures of celebrities and objects or remain silent if the name of the picture could not be produced. TOTs were identified in a subsequent session by re presenting the unnamed pictures and querying pa rticipants about the depictured person or object. As predicted, pictures of faces elicited more TOTs than pictures of objects. However, TOT rates for object
24 pictures were remarkably low (< 2%), suggesting that the non name stimuli may just have been too hi gh frequency to cause TOTs. Because the stimuli were not controlled for pre experimental familiarity, the increase in TOTs for proper names may have effect per se By u sing general knowledge questions to prompt TOT states as opposed to pictures, Burke et al. (1991) were able to use less frequent and more abstract non names to compare with proper names in their study. They found that older adults (but not young) had more TOTs for famous names relative to four other types of words (non object nouns, object noun, adjectives and verbs, and place names) i.e., a proper name effect However, the proper name effect only occurred when TOTs were computed as the proportion of total trials. When TOT incidence was computed as the proportion unsuccessful retrievals (i.e., # proper name effect disappeared. Computing TOTs as a proportion of unsuccessful retrievals is designed to correct fo r differences in item familiarity or level of difficulty (e.g., Brown, 1991). Thus, the lack of proper name effects for older adults using this measure may imply that the famous names used as targets may have been less familiar/more difficult than the othe r word types. However, it is worth noting that age differences in TOTs for proper names (where older adults experienced more TOTs than young adults) remained significant when the proportional TOT measure was used as the dependent variable. Recently, Hanle y (2011) conducted two experiments comparing correct response rates and TOT rates for proper names and non names matched for pre experimental familiarity. Converging with the evidence presented above, more TOTs occurred for
25 names relative to objects in Exp eriment 1, an effect attributed to a failure at phonological retrieval. However, object names produced significantly more incorrect responses, suggesting that participants were more inclined to supply an alternate word (e.g., producing flower for a picture of a corsage ) in the object naming condition relative to the face naming condition. Thus, the rate of TOTs for object names may have been support of this idea, Experiment 2 found no difference in TOT incidence between the using only words that had elicited no incorrect alternates during pilot testing). He concluded that names are not mor e difficult to retrieve; they merely lack feasible alternates, so proper name TOTs cannot be compensated by circumlocutions. It is worth noting that the two experiments in Hanley (2011) used different methods to induce TOTs (the first via pictures and the second via definition), adding an important additional variable that limits direct comparison between the two studies. As such, it seems premature to make such a strong conclusion about the source of proper name retrieval failures In sum, extensive resea rch suggests that representations of proper names and non names differ at the semantic and/or lexical level, a dissociation that causes increased resource burden during the retrieval of names. However, the specific locus of the distinction remains unclear. The few studies that have directly compared TOT rates for names and non names have differed in some fundamental ways: the mode of TOT elicitation (pictures or general knowledge questions), the features matched between proper name and non name stimuli, and the dependent measure used to assess TOT
26 incidence. The present research developed out of a need to confirm the existence of a name/non name retrieval disparity and to understand how and why this disparity increases with age The next section will address how current perspectives on proper name retrieval have been extended to explain why the difference between names and non names seems to be enhanced among older adults, which may provide further insight into the representation of names in memory. Aging an d Proper Name Retrieval A common theme inherent to explanations of the proper name/non name disparity and its interaction with age is that names are stored as more isolated representations within the mental lexicon, making them more sensitive to age relate d cognitive changes, brain injury, or disease pathology (e.g., Burke et al., 1991; MacKay & Burke, 1991; Semenza, 2006; 2009). Indeed, the subjective reports of older adults suggest that names are involved in more age linked retrieval failures than other t ypes of words diary studies examining real life TOTs among young and older adults (e.g., Burke et al., 1991; Cohen & Faulkner, 198 6 ) as well as a few laboratory studies comparing proper name versus non proper name retrieval among young and older participants (e.g., Burke et al., 1991; Cohen & Faulkner, 1986; Evrard, 2002; but see Maylor, 1995; 1997). Taken together, these findings support the idea of a disproportionate a ge impairment during the retrieval of proper names, such that age differences (between young and older adults) are larger for proper name s compared to non names, and/or that word class differences (between proper names and non names) are more pronounced fo r older adults than for young.
27 One possibility is that the independent processing stream dedicated to proper name retrieval is simply more resource demanding compared to other words, putting older adults (who have reduced cognitive resource capacity) at a distinct disadvantage. In his review, Sem en za (2006) described an unusual study where participants were asked to recall supraspan (i.e., exceeding typical short term memory capacity) lists of proper names and non names at sea level and at high altitudes, the latter representing a condition known to reduce cognitive function by causing hypoxia, deprived oxygen levels in the brain (Pelamatti, Pascatto, & Semenza, 200 3 ). Proper name recall was significantly impaired by the exposure to high altitude, whereas n on name recall was relatively resistant to the effects of hypoxia. Specifically, there was a significant reduction in the primacy effect during proper name recall under conditions of high altitude (memory for names at the beginning of the list), suggesting that semantic encoding of proper names into long term memory was most affected by the deficit in oxygen. Because proper name recall was particularly impaired by hypoxia, it suggests that the retrieval of names may be a more metabolically demanding task th an word recall (Semenza, 2006). Declines in general cognitive resources may contribute to age related changes in operate on the principle that the human capacity for p rocessing information is limited because a finite pool of resources is shared by competing mental processes that occur simultaneously or in close succession (e.g., Burke & Shafto, 2008). Because older g adults, they more consistently reach a point where available resources are insufficient to complete the
28 task. Specific resources implicated in age related language processes include processing speed, working memory, and inhibitory control. Evidence for a processing related changes to language (e.g., the speed with which language processes are being performed relative to manipu lations of other variables designed to increase task difficulty. Older adults also exhibit robust impairments in their ability to produce and comprehend syntactically complex language (e.g., Burke & Shafto, 2008; Kemper, 2006). Working memory declines prov ide the most intuitive explanation for changes in syntactic processing associated with aging. A syntactic structures, and as a result older adults produce sentences with lowe r syntactic complexity and propositional density (e.g., Kemper, 2006). Finally, the inhibitory deficit (ID) hypothesis (e.g., Hasher & Zacks, 19 8 8; Zacks & Hasher, 1994 ) suggests that older adults are less adept at suppressing irrelevant information from e ntering attentional awareness, causing increased word use and off topic verbosity during discourse production (e.g., Mortensen et al., 2006), and greater distraction from interfering stimuli during reading or listening (Burke & Shafto, 2008). In general, d eclines in processing speed, working memory, and inhibitory control effectively account for aging induced limitations on language processes that rely heavily on general cognitive resources. If proper name retrieval is particularly resource intensive, then it should show larger declines with age than non name word retrieval. What characteristics about the representation, organization, and processing of proper names demand more resources? The peculiar semantic qualities of names
29 described above (arbitrarines s, specificity, lack of meaning, and infrequency) may all contribute to why proper names are afforded less stable connections in the lexical semantic system and thus why proper names increase age differences in retrieval failures. In support of this idea, Fogler and James (2007) examined age differences in descriptive proper names. Older adults, but not young adults, recalled significantly more descriptive names (e.g., Pink Panther ) than nondescriptive na mes (e.g., Charlie Brown ) and were significantly faster in producing the descriptive names. Further, age differences in retrieval failures were larger for nondescriptive targets compared to descriptive targets. The authors argue that when a person views th e picture of a character with descriptive name (e.g., Pink Panther ), it automatically activates characteristics of the referent (e.g., being pink and a panther); if these characteristics are those that are also contained in the name, then the phonology of representation can be skipped. Additionally, the single connection constraint that is inherent for most proper names may not exist for descriptive names because they have additional connections between the node representing the name phrase and other semantic items and t he multiple connections could reduce age declines in retrieval. Overall, these findings suggest that name descriptiveness is one factor that influences age difference s in proper name retrieval. According to NST and its corollary TDH the particular difficulty with proper name retrieval among older adults is believed to be caused by the combination of two factors: 1) in the memory system, the lexical representations of proper names receive less activation than those of non names because proper names are not directly connected to
30 the semantic properties of the referent, and (2) with age, the cognitive system functions less well, and the units representing memory informat ion receive less activation (e.g., Burke et al., 1991). As discussed in the section on aging and lexical access, older relies on a single connection. Because the retrieva l of proper names seems to involve multiple isolated links (e.g., between the semantic system and name phrase node, the name phrase node and the individual first and last name nodes, and the name nodes to their corresponding phonological units), there is a greater likelihood that an age related transmission failure will result in a retrieval blockage. Although the subjective experiences of older adults and some empirical data lend support for the notion that aging exacerbates differences between proper nam e and non name retrieval, not all studies have found evidence for selective age impairments for proper names relative to other words. The interaction between aging and proper name/non name status seems to be contingent on the dependent measure used to asse ss word retrieval (e.g., James, 2006; Maylor, 1995; 1997). For example, James (2006) presented pictures of famous faces and asked young and older participants to name them as well as supply specific semantic information about the depicted targets. An analy sis of TOT incidence (the proportion of TOT responses in each condition) revealed an increase in age differences for proper names relative to the biograph ical information. However, age differences in non TOT retrieval errors (incorrect responses, unknown r esponses) were comparable for names and biographical information. Similarly, Maylor and Valentine (1992) asked participants aged 54 84 to complete a series of five tasks in response to picture of faces: structural decisions, familiarity
31 decisions, semantic decisions, first name decisions, and name retrieval. Regression analyses revealed linear age declines in all tasks except naming, which showed an exponential decline with age. Thus, the difference between proper names and other words may not emerge until the last stage of production, i.e. phonological retrieval, the stage of production wherein older adults have consistently shown exacerbated declines. to pictures of fam ous faces were categorized as follows: unfamiliar, familiar (but no semantic information available), semantic information (knowledge of biographic facts but no name), and name (proper name given with semantic information). She then computed conditional pro babilities that were used to assess the likelihood of retrieving a particular piece of information contingent upon having gotten to the previous stage (e.g., accessing semantics after recognizing the face, accessing the name after activating semantics). Us ing these conditional probabilities, she reported no exceptional age deficit at the level of name retrieval relative to the earlier stages of face recognition and semantic retrieval. However, participants in these analyses were in their 50s, 60s, and 70s, making it difficult to make comparisons to other studies that use college students as their young adult group. Nonetheless, the mixed results across studies highlight the importance of considering the type of dependent measures used to assess retrieval suc cess prior to making global claim about age related declines So far, the majority of research on proper name retrieval and its decline with age has focused on the semantics of name retrieval, a non intuitive fact considering that the naming failure stem s from a problem during phonological retrieval. Additionally, it is well established that older adults have greater difficulty with phonological level
32 processes compared to semantic level processes, suggesting that a selective age deficit for names may be caused by characteristics of names that emerge at the phonological level, or the links to it. One possibility is that names consist of lower frequency phonological components relative to other words, hindering phonological encoding. According to the phonol ogical plausibility hypothesis, names are less restricted by the phonological constraints imposed by a language, so there are a greater number of plausible phonological combinations used in names compared to other words (e.g., Brennen, 1993). Due to the re laxed phonological constraints, names would be more likely to possess very low frequency phonological units. According to Brennan (1993), we are much more likely to encounter rare or novel syllables when we learn a new name compared to when we learn new wo rds, therefore making it more difficult to encode the new name and integrate it with existing items in the phonological lexicon. Nearly 80% of English words contain the 500 most frequent syllables (e.g., Levelt et al., 1999), so phonological encoding and m otor planning procedures are better practiced for high frequency (HF) syllables compared to low frequency (LF) syllables (e.g., Cholin, Dell, & Levelt, 2011; Farrell & Abrams, 2011). According to interactive activation models of speech production, words wi th HF syllables have stronger and more stable connections between the lexical and phonological levels due to the repeated activation of regularly accessed phonological representations. Each time a syllable is activated, it spreads bottom up activation to a ll lemmas sharing that syllables. Over time, HF syllables acquire stronger connections to the lexical layer due to the repeated spreading of activation between the lemmas that contain that syllable. If proper names contain LF phonological forms, they may h ave weaker lexical to phonological connections because
33 this bottom up strengthening activity does not occur very often. This assumption would explain why proper names induce more TOTs than other words and why this difference increases with age. Older adult proper names. Conversely, the difficulty with names may be caused the opposite phenomenon: too much sharing name is typically unique, most people possess a first or last name that is common among many others, which may cause inhibition during retrieval (e.g., Maylor, 199 7 ; Valentine et al., 1996) This disse rtation aimed to disambiguate the unique contributions of phonological and semantic factors in the disproportionate impairment for proper name retrieval in aging. To elucidate structural differences in the semantic and phonological representations of prope r names and non names, phonological and semantic primes were introduced during different stages of the TOT experience, an implicit way of measuring interactivity between words and names. Priming and Tip of the Tongue (TOT) States The application of primi ng methods to the TOT phenomenon has provided invaluable insight into the time course of lexical access and the level at which retrieval failures occur. Both phonological and semantic primes have been investigated within the TOT literature, with the logic being that each type of prime should be differentially useful depending on when they are encountered during the production process and where the blockage has occurred. Beyond TOTs, research on phonological priming using a variety of speech production tasks has shown that the presentation of a phonologically related prime word facilitates production, an effect attributed to the level
34 of phonological encoding (e.g., Abrams, White, & Eitel, 2003; Cutting & Ferreira, 1999; Damian & Martin, 1999; James & Burke, 2000; Meyer & Schriefers, 1991; Schriefers, Meyer, & Levelt, 1990; White & Abrams, 2002 ). Prior exposure to a prime word ( idol ) activates the phonology it shares with the target ( island ), therefore speeding the time required to retrieve the first syllable of the target for production. Specific to the TOT literature, the implicit presentation of a phonologically related prime has been shown to reduce the incidence of TOT states as well has increase resolution once a TOT state has occurred (e.g., Abrams & Ro driguez, 2005; Abrams et al. 2003; James & Burke, 2000; Meyer & Bock, 1992; White & Abrams, 2002). Because TOTs occur when some or all of the phonological components of an intended word are temporarily inaccessible, the presentation of phonological prime is presumed to send activation to the blocked phonology through an alternate route, thereby attenuating the occurrence of TOTs and facilitating increased retrievals after TOTs have occurred. Despite weakened links to phonological forms, older adults in the ir 60s and early 70s (i.e., young old adults) experience reduced TOT incidence and boosts in TOT resolution due to phonological priming (e.g., Abrams, Trunk, & Merrill, 2007; James & Burke, 2000; White & Abrams, 2002). Differences in the size of priming ef fects between young and young old adults have been inconsistent and unreliable, with young adults experiencing more priming in some studies (e.g., Abrams et al., 2007) but not in others (e.g., James & Burke, 2000; White & Abrams, 2002) However, there seem s to be a limit on the extent to which transmission deficits can be offset by priming. TOT resolution in older adults in their mid 70s and older (those with the most extensive declines in connection strength) is not
35 affected by priming, at least with a sin gle initial syllable (Abrams et al., 2007; White & Abrams, 2002). Phonological priming has recently been shown to influence the incidence (e.g., Burke et al., 2004) and resolution (e.g., White, Abrams, & Frame, in press) of proper name TOTs. Burke et al. ( 2004) found that participants experienced fewer TOTs for proper name target pictures (e.g., Brad Pitt ) when they had recently produced a pit ) in response to a definition compared to an unrelated word (e.g., cane ). Simi larly, White et al. (in press) reported phonological priming of TOT resolution when TOTs for famous proper names were elicited via general knowledge questions (e.g., What actor screenwriter directed and starred in the film Annie Hall? Target = Woody Allen ). Participants were more likely to resolve their TOTs when the subsequent question contained a prime with the same first name as the target (e.g., Woody Harrelson ). Although TOTs for both proper names and non names benefit from phonological priming, prim ing effects are moderated by specific characteristics of the prime, including grammatical class and first syllable frequency (for non names), and semantic category (for names). Among non names, phonological primes facilitate TOT resolution only when the pr ime is a part of speech different from the target (Abrams & Rodriguez, 2005; Abrams et al., 2007). At the level of phonology, Farrell and Abrams (2011) found that the size of phonological priming effects was contingent on the frequency of the first syllabl e shared between target and prime. Larger priming effects on TOT resolution were observed when the target had a LF first syllable compared to a HF first syllable, a pattern that was constant across age groups. Both effects have been attributed to
36 interfere nce from phonological competitors, even after initial lexical selection has been complet ed Because similar sounding words from the same grammatical class might be viable candidates for selection, they compete with the TOT target and forestall target retri eval. Similarly, syllable frequency moderates the size of priming effects by determining the number of partially activated lexical candidates. Because HF syllables occur in many other words, priming is diffusely distributed among more alternate words, ther efore reducing the priming effect for targets with HF syllables compared to LF syllables. At present, it remains unclear whether phonological primes exert analogous effects on proper names, although recent research suggests that names are also susceptible to competition from similar sounding names when they are viable candidates for selection. White et al. (in press) found that TOT resolution for a famous target ( Elton John a singer ) was marginally higher following the presentation of a prime with the sam e first but only when the prime was from a different semantic category ( Elmer Fudd a cartoon character ). First syllable primes had no effect on TOT resolution when the prime was from the s ame semantic category ( Elvis Presley a singer ). In contrast, primes that shared an entire first name as the target (described above) universally boosted TOT resolution regardless of semantic category These results were taken to suggest that the phonolog ical input from a single syllable is insufficient to prime target retrieval when there is also competition at the semantic level. It is interesting to note that priming effects from first syllable primes were only marginal, even for different semantic cate gory primes. Relative to non names, it may be more difficult for a single syllable to
37 send enough activation to a proper name target to facilitate its retrieval due to the sheer number of connections engaged during proper name retrieval. As shown in Figure 1, there are connections between the name phrase and the individual first and last name nodes, then numerous connections between the first/last name nodes and their phonological subcomponents. In contrast, TOTs for a word would only involve the connection s between the lemma and its phonological subcomponents Alternatively, if names contain lower frequency phonological units compared to other words, they may show larger priming effects because priming will spread to fewer phonologically related candidates In support of this idea, one of the case studies names when given the first phoneme of a proper name, but not when given the first phoneme of non proper nouns. Semenza arg ued that proper names are better able to benefit from the phoneme cueing because of the one to one connection between the proper name representation and its phonology, i.e., less opportunity for competition. In sum there is evidence to suggest that proper names may require more phonology to show priming effects due to increased competition from similar sounding names and the greater number of steps involved during retrieval which may weaken priming over time. Conversely, names may benefit more from primin g due their possession of infrequent phonology, which would allow for priming to be distributed among fewer lexical candidates. Research investigating semantic priming effects on TOTs and correct word retrieval is less consistent in that semantic primes can be facilitative or inhibitory depending on the type of task. Among non names, some production tasks, such as
38 word naming, have shown facilitation from semantic priming presumed to be caused by spreading activation between conceptually related words, al lowing for heightened semantic activation of a target (see Farrell, Abrams, & White, 2012, for a review). In contrast, semantic interference has been reported in other production paradigms, such as picture word interference where naming a target picture ( e.g., lion ) is slowed by the presence of a semantic associate word (e.g., tiger ) presented prior to or simultaneously with the target picture (Cutting & Ferreira, 1999; Sailor, Brooks, Bruening, Seiger Gardner, & Guterman, 2008; Schriefers et al., 1990; St arreveld & La Heij, 1995, 1996). Theoretically, interference from a semantically related alternate word would occur if it initiates competition at lexical selection, thereby slowing or preventing the selection of the correct target, while semantic facilita tion would occur when there is spreading activation within the semantic system that does not result in the activation of lexical competitors. Because TOTs are thought to be caused by failed phonological retrieval following successful lexical activation, se mantic primes should not impact TOT states but might exert an influence on correct target retrieval. In support of this idea, a few studies have shown that semantically related cues presented prior to, or concurrently with a TOT question, d id not affect t he incidence of TOTs relative to unrelated words (e.g., Jones, 1989; Meyer & Bock, 1992, Experiment 1) but increased the proportion of correct responses (Meyer & Bock, 1992, Experiment 1 and Experiment 2). However, one study ( Meyer & Bock, 1992, Experiment 2) found that semantically related cues increased TOT responses compared to unrelated words when presented after an initial response failure It is important to note that in this methodology participants were told that the cues might provide some informat ion about
39 the target word, so the results represent explicit search strategies as opposed to implicit spreading activation from semantically related words. Therefore, it is unclear whether increased TOTs in the semantic cue condition resulted from particip ants becoming fixated on the cue itself (therefore hindering target retrieval) or from automatic processes like spreading activation Nonetheless, evidence for increased TOTs from semantic cues necessitates the need for further research on the role of sema ntic competitors in causing TOTs and in preventing TOT resolution. Finding semantic interference from implicitly presented semantic primes would suggest that there may multiple sources of retrieval failures that result in TOTs, such as incomplete activatio n of An interesting empirical question is whether proper name retrieval is similarly susceptible to semantic interference and/or facilitation as is non name retrieval. Because of the sema ntic irregularities associated with proper names, there is reason to suspect that proper names may behave differently when presented with a semantic prime than do non names. For example, Vitkovitch, Potton, Bakogianni, and Kinch (2006) found that participa nts produced fewer naming errors and time out responses when naming pictures of famous faces (e.g., Nicole Kidman an actress ) if they had named a semantically related face three trials previously (e.g., Julia Roberts an actress ). In Experiment 1, primes and targets were from the same occupational category but were not often linked to one another. However, the facilitation effect held in Experiment 2 when the prime and targets were also semantic associates (e.g., Jamie Oliver Gordon Ramsey celebrity chefs with television shows ) and in Experiment 3 when the prime was presented in word form instead of a picture. Thus, the authors concluded
40 that semantic lexical processing of proper names differs from non names in that it involves little or no competition at the lexical level. O nly one published study has investigated the influence of purely semantic prim es on proper name TOTs specifically. Cross and Burke (2004) found that the incidence of TOTs for famous names (e.g., Audrey Hepburn ) was not affected by prio r production of a famous character that the target has played (e.g., Eliza Doolittle ). However, the character priming manipulation did reduce the proportion of incorrect responses. This pattern suggests that spreading activation within the semantic network lexical representation, thereby enabling more correct responses. However, because TOTs are caused by a breakdown in the lemma to phonology link, semantic priming had n o effect on this later stage in the production process. In sum, semantic and phonological primes exert differential influences on word retrieval depending on when they are encountered during production. As yet, no studies have directly compared the effec tiveness of priming on proper names and non names, which may serve as a useful tool in establishing how these word types differ in terms of their representations at the semantic and phonological levels. The Current Research How names are differentially r epresented from non names and whether this separation renders name retrieval more difficult compared to other words remains a subject of empirical and theoretical contention. Moreover, the question of whether older adults have disproportionate difficulty d uring name retrieval is even more complex and inconsistent. This dissertation aimed to clarify these issues by directly comparing the two word types within a single paradigm and by controlling as many features of the
41 stimuli as possible. The first phase of the project consisted of a pilot study to assess name stimuli and to assess semantic similarity among target prime pairs. The ultimate goals of the project were to examine th e extent to which the disproportionate age deficit for names is caused by semantic level factors (as proposed by many theoretical perspectives), phonological level factors (as proposed by the current author), or a combination of the two. To accomplish this end, two TOT experiments were conducted to compare the influence of phonological (Experiment 1) and semantic (Experiment 2) primes on the incidence and resolution of TOT states for names and non names. The influence of both types of primes differed as a f unction of age and target type, revealing structural differences in the retrieval pathways dedicated to proper name and non name retrieval and links that are most vulnerable to aging. Further, the retrieval of both types of targets were strongly linked to first syllable frequency, providing novel insights into the lexical to phonological connections of proper names and other words and how these connections change throughout the lifespan.
42 John Baker (Noun Phrase) Joh n B) Semantic System Phonological System Wears Hat Makes Bread Baker (Occupation Noun ) ( BA KER S yllable Nodes Semantic System Visual Concept Node for John Baker Cooks Well Plays Tennis Has a brother Baker BA KER Syllable Nodes Smell like cookies A) Figure 1 1 Representation of Proper Names and Non Names according to NST (adapted from MacKay & Burke, 1991). Visual Concept Node for John Baker
43 CHAPT ER 2 PILOT STUDY Method Data Collection and Participants Participants consisted of 18 young adults ( M = 19.32, SD = 1.25, range = 18 22 years) and 18 older adults ( M = 77.4, SD = 8.8, range = 60 92 years). Young adults were recruited from the Psychology Re search Participation pool at the University of Florida. Older adults were primarily recruited from the Cognition and Aging Lab Older Adult Pool, a database of community dwelling volunteers living in Gainesville and surrounding areas. Materials One hundred fifty non names and 150 famous proper names were developed for the pilot study. Because the ability to conduct TOT resolution analyses is reliant upon the number of TOTs participants experience during an experimental testing session, 47 proper names and 7 2 non names known to induce high rates of TOTs based on previous research in our laboratory, were selected. One hundred three new proper names and 78 non names were also developed in an attempt to find more items that were potentially susceptible to TOTs and were reasonably familiar to each age group. Non name and proper name targets were selected from a variety of semantic categories, described in greater detail in the experimental methods. Each target was paired with (1) a phonological prime word/name, (2) a semantic prime word/name, and (3) an unrelated word/name. For the non name targets, the phonological prime word contained the same phonological first syllable as the target, as determined by an online dictionary ( http://www.merriam webster.com/ ), and was
44 unrelated to the target in meaning. Previous TOT research has demonstrated that phonological priming is only effective when the prime is from a different part of speech as the target (Abrams & Rodriguez, 2005; Abrams et al. 2007). Therefore, I selected phonological primes that were from a different grammatical class as the target (e.g., a verb, adjective, adverb, interjection) in order to maximize phonological priming effectiveness. S emantic primes were synonyms for the target word, superordinate/subordinate members of the same semantic category, or words that share defining semantic features with the target. To maximize the semantic similarity between targets and primes, semantic primes were the same par t of speech as the target (both were nouns). The unrelated words, half of which were nouns and half of which were non nouns, were semantically and phonologically unrelated to the target. ith the same ., Al bert Einstein for Al fred Hitchock ). Phonological prime names and the unrelated names were the same gender as the target but differed in terms of occupational category and visual similarity in order to eliminate any semantic relation to the target. The semantic primes came from the same occupational category as the target and were the same gender Semantic primes for non fictional targets were also the same race/ethnicity as the target with the exception of three targets Target names that are fictional characters were paired with semantic primes that are also fictional characters within the same modality. More details and examples for both types of targets and their phonological, semantic, and u nrelated primes are provided in the experimental method.
45 Procedure The pilot study consisted of two major subsections. In the first subsection, participants were asked to rate their knowledge and familiarity with half (75) of the non name target words and half (75) of the proper name targets. In the second subsection, participants were asked to make semantic similarity judgments between the remaining 75 non names targets and either their semantic, phonological, or unrelated prime and between the remaining 7 5 proper name targets and either their semantic, phonological, or unrelated prime. As such, participants saw each target only once and either rated primes. The pilot study began with the non name familiarity rating task. Participants were told that they were going to be asked to rate their familiarity with famous names and uncommon words. To discourage the inflation of familiarity ratings due to participant response b ias, the participants were told that the names and words were developed for a variety of age groups and were not expected to know them all. They were then at a time. You r task is to think about how familiar you are with each word. For each item, decide how well you know the word on a scale of 1 (completely unfamiliar) to 5 point Likert scale with descr iptions for each numeric option was presented below the instructions ( Figure 2 1). Once participants indicated that they understood the task and scale, the experimenter pressed the Space Bar key to begin the study. Each trial was preceded by a fixation cro ss displayed in the center of the screen, and the experimenter pressed Enter to begin the trial. Targets were displayed one at a time and were centered in the
46 upper one third of the screen. The Likert scale was displayed below the target during each trial. Participants were asked to make their selections out loud, and the experimenter recorded their responses via key press (i.e., 1, 2, 3, 4, or 5) After going through all non name targets, participants were then presented with the following instructions: For each name, decide how familiar you are with that person. When determining familiarity, think about how many facts you know about each person and how regularly you encounter the name. Use the f scale was developed for the proper name targets to assist participants in assessing their familiarity with famous names ( Figure 2 2). The same procedure used for non name familiarity ratings was used for th e 75 proper name targets, with proper name targets displayed above the scale. After participants completed the familiarity ratings, they were then asked to make judgments about the semantic similarities between the non name targets and their primes. Each target that was not rated for familiarity during the first subsection was presented in conjunction with its semantic, phonological, or unrelated prime word. Using a 3 point Likert scale, participants were asked to assess how closely the target and primes w ere related to each other in meaning ( Figure 2 3). Participants were given the following instructions for the non pairs of words. For each pair, think about how similar in meaning the two words are to one another, and then rate their similarity using the scale below. It may help to think about how many 'features' the two words have in common. For example, the words 'dog' and 'cat' are STRONGLY RELATED because they are both animals, mammals
47 with fur, and do mestic pets, and you would say 3. 'Dog' is SOMEWHAT RELATED to 'snake' because they are both animals (you would say 2), but COMPLETELY UNRELATED to the word 'boat' because they share no common features (you would say 1). Although some of the words may soun meanings when determining your rating. If you do not know the meanings of one or both understood the similarity ratings task, the exper imenter pressed Enter to begin the next subsection. As with the familiarity ratings, each trial began with a fixation cross to ready the participants. The target words were displayed directly above the prime words, and the word pairs were centered approxim ately one third of the way down from the top of the screen. The scale appeared below the word pairs on each trial, and the After going through all 75 non name targets, participants performed the same task this section, you will see pairs of famous names. Your task is to decide how similar the two people are to one another in terms of their biogra phical characteristics (e.g., age, era, occupation), personality features (funny, mysterious), and public personas (e.g., political enemy, villain). For example, Robert DeNiro and Al Pacino are STRONGLY RELATED because they are both actors, of Italian desc ent and have played tough guys in Mobster movies, so you would say 3. DeNiro would be SOMEWHAT RELATED to Brad Pitt because they are both actors (you would say 2), but COMPLETELY UNRELATED to Benjamin Franklin because they lived in different eras, held dif ferent occupations, and have dissimilar personas (you would say 1). If you don't know one or
48 both names, select 9 (unknown). Note that all pairs contain to two people from the same gender, so gender should not be used as a criteria to judge similarity. Alt hough some of the names may sound similar to one another, think about the characteristics of the semantic relatedness is thought to differ for names, a modified scale was develop ed for the proper name similarity ratings task ( Figure 2 4). Results The pilot test had the following specific aims: (1) to find a subset of targets that were sufficiently familiar to both young and older adults, (2) to address potential age differences i semantic primes were more closely related to their targets than either the phonological or unrelated prime words. To address the first goal, a 2 (Age Group: Young, Old er ) X 2 (Target Type: Non Name, Name) mixed factorial ANOVA was conducted by items on the mean familiarity ratings. Results revealed that older adults ( M = 4.1) rated the targets as more familiar overall than young adults ( M = 2.9), F (1, 298) = 430.2, MSE = .528 p < .001, and that non names ( M = 3.95) were rated as more familiar than names ( M = 3.0), F (1, 298) = 178.44, MSE = .752, p < .001. The Age Group X Target Type interaction was not significant, F < 1, p = .528. Although the lack of interaction revealed t hat both age groups were less familiar with the proper name targets, the primary point imp ortant objective was to identify a subset of items that would be moderately familiar to each age group (Aim 1), the first step was to remove the proper names targets with the lowest familiarity ratings as well as
49 responses during the semantic similarity ratings task. The primary focus was to remove items that were especially unfamiliar to young adults, in an attempt to reduce age differences in familiarity (Aim 2). Finally, I removed items where the semantic simil arity ratings were not sufficiently higher in the semantic prime condition relative to the phonological or unrelated prime, thereby indicating a weak semantic relation to the semantic prime and/or low familiarity with the targets and primes (Aim 3). The r emoval of these items resulted in 76 non name targe ts and 76 proper name targets ( Table 2 1 for descriptive statistics on the familiarity and similarity ratings for name target has a corresponding proper name target with the same first syllable. A 2 (Age Group: Young, Older) X 2 (Target Type: Proper Name, Non Name) mixed factorial ANOVA by items was used compare the mean familiarity ratings for proper name and non name tar gets among young and older adults. Results revealed a significant main effect of target type, F (1, 150) = 83.2, MSE = .338, p < .001, where non name targets ( M = 4.0) were rated as more familiar than proper name targets ( M = 3.5). There was also a signifi cant main effect of age group F (1, 150) = 215.09, MSE = .469, p < .001, where older adults ( M = 4.35) were significantly more familiar with targets than young ( M = 3.2). The Age Group X Target T ype interaction was not significant, F < 1, p > .457. Althoug h the pattern was similar to the analysis on all stimuli, the removal of items resulted in a meaningful increase in familiarity with each target type, such that now both T he lack of an interaction between age and target type is especially important for this study because any disproportionate age differences in TOTs for names versus non
50 names cannot be attributed to familiarity discrepancies; in other words, older adults are more familiar with both word types than young adults, and to the same extent. To assess whether the semantic primes were given higher semantic similarity ratings than the phonological and unrelated primes for young and older adults, a 2 (Age Group) X 3 ( Prime Type: Semantic, Phonological, Unrelated) mixed factorial ANOVA was conducted by items on the mean semantic similarity ratings for each item. Results revealed a significant main effect of prime type, F (2, 240) = 1131 .29 MSE = .14, p < .001, where se mantic primes ( M = 2.57) were rated as more similar to the target than either the phonological ( M = 1.16) or unrelated primes ( M = 1.17), which did not differ. There was also a significant main effect of age group F (1, 120) = 5.18, MSE = .105, p = .025, where young adults ( M = 1.66) rated the primes as more semantically similar to the target than did older adults ( M = 1.6). There was a significant Age Group X Prime Type interaction, F (2, 240) = 13.624, MSE = .099, p < .001, such that young adults provide d higher ratings of similarity than older adults for the phonological and unrelated primes ( p s <.001), but lower ratings of similarity for the semantic primes ( p = .042). This may reflect the fact that young adults were less certain of the meanings of the phonological and unrelated primes so were more liberal with their ratings of similarity to semantic primes as more similar than either the phonological or unrelated primes ( p s < .001), which did not differ from one another ( p s> .264).
51 Table 2 1. Means and standard deviations for familiarity ratings and semantic similarity ratings in the pilot study Age Group Young Old er Non Name Targets Familiarity Rating 3.5 (.65) 4.6 ( .32) 1.2 (.36) 1.1 (.19) 2.6 (.4 1 ) 2.7 (.35) 1.2 (.3) 1.1 (.17) Proper Name Targets Familiarity Rating 2.8 (.78) 4.1 (.69) Phonological Prim 1.2 (.34) 1.1 (.25) 2. 4 (.58) 2.5 (.47) 1.4 (.46) 1.1 (.19) *Note: Familiarity Ratings are based on a 5 point scale (1= Completely Unfamiliar, 5= Very Familiar). Semantic similarity ratings are based on a 3 point scale (1= Completely Unrelated, 3= Strongly Related)
52 1 2 3 4 5 Completely Unfamiliar Somewhat Unfamiliar Vaguely Familiar Familiar Very Familiar I have never heard this word and could not define it or describe anything about it I have heard the word but could not define it or use it in a sentence I have heard the word and produced it before. I could not give a definition or use it in a sentence, but I may be able to select the correct definition if given choices I hear and use the word occasionally. I could use the word correctly in a sentence and could easily define it I hear and use the word regularly and could provide a clear definition of the word Figure 2 1. Familiarity rating scale for non na me targets 1 2 3 4 5 Completely Unfamiliar Somewhat Unfamiliar Vaguely Familiar Familiar Very Familiar I have never heard of this person and could not tell you a single fact about him or her I think I have heard this name before, but I am fairly sure t hat I have never talked about this person. I could not tell you anything about him/her with any confidence I have heard or talked about this person before, but it would be difficult to tell you more than a single fact about him or her I hear and talk about this person occasionally. I could probably tell you a couple of facts about him or her I talk about this person often and could tell you many facts about him or her Figure 2 2. Familiarity rating scale for proper name targets
53 Figure 2 3. Semantic similarity ra ting scale for non name targets Figure 2 4. Semantic similarity rating scale for proper name targets 1 2 3 9 Completely Unrel ated Somewhat Related Strongly Related Unknown The two words share no common features and are not similar in meaning The words' meanings share at least one common characteristic. For example, they are from the same basic category: animal, object, tool, abstract concept, food, plant, etc. The words come from the same specific category, are united by a common concept, have multiple shared characteristics, or are synonyms I do not know the meaning of one or both words 1 2 3 9 Completely Unrelated Somewhat Related Strongly Related Unknown Besides gender, the two people do not have a single feature in common The two people have the same occupation, and I could probably name 1 or 2 add itional features the two people have in common The two people have the same occupation and age range, and I could name 3 or more additional features they have in common I am unfamiliar with one or both names
54 CHAPTER 3 EXPERIMENT 1: PHONOL OGICAL INFLUENCES ON NON NAME AND PROPER NAME TOTS Specific Aims Specific Aim 1 To examine the role of phonological frequency in the disproportionate impairment for proper name retrieval and its interaction with age. Only a handful of studies have directly compared the retrieval of names and non names in single laboratory study and the heterogeneity of methods and dependent measures have made it difficult to draw strong conclusions about the pre sumed disparity between name and non name retrieval. As such, the fundamental goals of the study were to confirm the distinction between proper names and non names in terms of retrieval difficulty and to investigate how differences between word types are a ffected by aging. Assuming that names induce more TOTs than non names, especially for older adults, the second goal of this project was to clarify the characteristics of names that make them more susceptible to retrieval failures. Of particular interest wa s how phonological frequency (i.e., first syllable frequency) would moderate the relationship frequency phonological forms might make them particularly vulnerable to age r elated retrieval failures, accounting for the disproportionate age related deficit for proper names. An alternative possibility was that similar sounding names might interfere with name retrieval due to the sharing of first names and surnames which would cause names with many first syllable neighbors (HF first syllables) to be more susceptible to TOTs
55 Specific Aim 2 To investigate how phonological priming may attenuate the proper name/non name retrieval disparity and its increase with age. Because TOTs are thought to be caused by insufficient activation of an intended reduce the proportion of TOTs overall. Previous exposure to the full phonology of a target has been shown to reduce TOTs and increase correct responses for both names (Burke et al., 2004) and non name targets (James & Burke, 2000). However, the present study was the first to directly compare the effectiveness of phonological priming for proper names versus non names using identical methods. It was also the first to assess phonological priming effects on TOT incidence for proper names using only the There were a priori reasons to suspect that proper names would benefit more from priming due to the solitary connections between semantic/lexical and phonological representations. Conversely, proper names may require more phonological input than a single syllable to show robust prim ing effects (e.g., White et al., in press). Priming was expected to reduce age differences in TOT rates by compensating for age related transmission deficits to phonological forms. It was unclear how phonological priming from only a single syllable would influence the proportion of correct responses, a less direct indicator of phonological activation than the proportion of TOT states. Theoretically, the rate of correct responding would be linked more strongly to successful spreading activation from the sem antic level to the lexical representation of the correct target, and as such should be less influenced by phonological priming than TOTs. Although previous research has
56 shown that phonological priming can increase the proportion of correct responses (Burke et al., 2004; James & Burke, 2000), both studies presented the entire phonology of a target either through the use of homophones, in the former case, or multiple prime words, in the latter case. Thus, providing the first syllable of the target through imp licit priming may or may not provide enough bottom up activation to exert an influence on lexical selection and the proportion of correct responses. It was predicted that phonological primes presented before the target question would reduce TOTs (for which the activation of the first syllable is critical) but would have minimal influences on the proportion of correct responses (which is more dependent on earlier processes such as lexical selection). Specific Aim 3 To evaluate the effectiveness of phonologi cal priming on successful TOT resolution syllable frequency. Previous research has demonstrated that encountering the phonology of a forgotten word is more helpful at facilitatin g TOT resolution when the word begins with a LF first syllable compared to a HF first syllable (Farrell & Abrams, 2011). Presumably, priming is less beneficial for words with HF phonological forms because there are many alternate words to compete with the target at the lexical level, therefore reducing the facilitation effects occurring at the phonological level. If this effect should hold, priming of TOT resolution should be more effective for proper names than non names because there are fewer individual words sharing phonology (e.g., Brennen, 1993) and thus less opportunity for lexical competition. Conversely, because the lexical phonological representations of proper names require the activation of multiple isolated links, phonological priming from a sin gle syllable may be less useful in facilitating TOT
57 resolution for names compared to non names. The influence of priming on phonological and lexical level processes was not expected to differ as a function of age, as the prime word should effectively offse t the age related decline in phonological activation Method Participants Young adults (Range = 18 28 years) were recruited from various sources in the Gainesville area, including University of Florida courses, local fliers, and advertisements placed on i nternet classifieds. Participants from the University of Florida Undergraduate Psychology Participant pool or the Undergraduate Linguistics Participant pool received partial course credit for participation, and individuals recruited via outside sources rec eived $10 compensation for their time. Older adults (Range = 60 75 years) were also recruited using a variety of methods, including the existing Cognition and Aging Laboratory participant pool database, email solicitations to University of Florida alumni, and advertisements at retirement communities and senior centers in Gainesville, Florida, and Fort Myers, Florida. Older adults were compensated at a rate of $8/hour. All participants were native English speakers with normal or corrected to normal vision an d hearing and no diagnosis of a learning disability or cognitive impairment. Prior to completing the experiment, both age groups were asked to complete a brief demographic questionnaire to collect information about their health and educational history. Thi s step was particularly important for older adults in order to rule out any existing conditions (e.g., stroke) that may impact cognitive performance. Older adults then completed the Mini Mental Status Exam (MMSE; Folstein, Folstein, & McHugh, 1975) as a cu rsory screen for dementia and were required to score 25 or above to be included in the study.
58 One older adult participant was removed from analyses because of a low MMSE score, resulting in 62 young adults ( M = 19.4, SD =1.6) and 43 older adult ( M =68.2, SD = 4.7) participants. Young adults were evenly represented by males and females. Older adults consisted of 35% males and 65% females. Full descriptive statistics in Tabl e 3 1. Independent samples t tests were used to compare the two age groups on the following dimensions: perceived health rating relative to others at same age (on a 10 point scale), years of education, vocabulary ( 25 item multiple choice test requiring par ticipants to select the nearest synonym to an uncommon word), forward digit span (the number of digits that could be recited back to the experimenter in the same order), and backward digit span (the number of digits that could be recited back to the experi menter in reverse order). Relative to the young adult group, older adults had more years of education, higher vocabulary scores, and shorter forward digit spans ( p s <.001). The age groups did not differ on their ratings of health or backward digit span ( p s > .261). Materials Seventy six proper names and 76 uncommon non name nouns were selected syllable frequency was provided by the CELEX database (Baayen, Piepenbr ock, & Gulikers, 1995), the only normative corpus that provides syllable frequency measures in English. First syllable frequency was defined as the summed frequency of words containing a particular syllable in the onset position. For the proper name target s, first syllable frequency was based on the first syllable of the first name. Although the corpus includes few proper names, it was used to ascertain the frequency of each proper name
59 y of use among existing non names within the database. If the syllable did not appear in the CELEX database, it was assigned a frequency value of 0. Half of the targets were categorized as having a HF first syllable and half were categorized as having a LF first syllable, as determined by a median split within each target type ( median = 498 for non names, 361 for names). Example non name and proper name targets with HF and LF first syllables are shown in Table 3 2 and Table 3 3 along with the mean syllable frequency for each category. Importantly, there was a subset of stimuli for which syllable frequency did not differ between non names and proper names because they possessed the same first syllable : Half of the non name targets had a proper name counterpa rt with the same first syllable (e.g., Audiologist/Audrey Hepburn ; Diameter/Diane Sawyer ). Non name targets were nouns from a variety of semantic categories, including objects ( kaleidoscope ), abstract words ( nostalgia ), procedures/processes ( photosynthesis ), people ( equestrian ), literary terms ( acronym ), and actions ( arson ) Proper name targets were also selected from a variety of occupational categories, including actors ( Eddie Murphy ), living politicians and world leaders ( Rudy Giuliani ), historical figur es/presidents ( Herbert Hoover ), artists/writers/directors/musicians ( Andy Warhol ), athletes ( Michael Phelps ), fictional characters ( Betty Boop ), entrepreneurs and name brands ( Orville Redenbacher ( Monica Le winsky ). Among the 76 proper name targets, 52 were male and 24 were female. T argets were also selected from a variety of time periods, although the distribution favored historical figures and celebrities born prior to the young adult age range. Nineteen ta rgets were born prior to 1900, 11 were born between 1900 and 1919,
60 15 were born in the 1920s and 1930s, 13 were born in the 1940s and 1950s, 12 were born in the 1960s and 1970s, and 6 were born in 1980 or later. Each target was paired with a phonological prime word and an unrelated word. For non names, the phonological prime had th e same first syllable as the target but was from a different grammatical class (26 verbs, 49 adjectives, and 1 adverb). The unrelated words were a mix of nouns (39) and non nouns (12 verbs, 25 adjectives) and were used in both Experiment 1 and Experiment 2. The unrelated words had no phonological or semantic relation to the target. For proper names, the phonological rst name. Both the phonological prime name and the unrelated name were the same gender as the target but came from a different occupational category in order to minimize the semantic relation to the target ( Table 3 2 and Table 3 3 for example targets and p rimes) A trivia style question was developed for each target that was designed to elicit the target as the answer. For non name targets, the question asked the participants to provide the specific word that embodies a definition (e.g., What do you call a word or abbreviation, such as CEO, that is formed by taking the first letter or letters from each word in a phrase? Target = acronym). Target questions were written to include enough specific details so that there was only one correct answer to the questi on (i.e., the target). This task proved very difficult for some of the non names targets that had very close synonyms or semantic associates (e.g., tyrant, christening). In these cases, we inserted a phrase to discourage the production of an alternate word (e.g., ). For proper name targets, the questions asked the participant to supply the name of a famous person based on biographical information
61 (e.g., Which English filmmaker was the master of suspense, making film s such as Psycho and Vertigo? Target= Alfred Hitchcock). At least two to three biographic details were embedded in the proper name target questions to account for the fact that young and older participants may have been exposed to the famous names in diffe rent contexts. The non name and proper name target questions were controlled for word length so that differences in comprehension difficulty could not account for effects of target type (refer to Table 3 4 for mean word length of all questions). A one way ANOVA confirmed that non name target questions and proper name target questions did not differ in word length, F ( 1 150 ) = 1.37 MSE = 35.61, p = 244 Importantly, contain ed within the words in the questions. The phonological prime and unrelated words were embedded in questions that were presented before or after the target questions. The answers to the prime/unrelated questions were phonologically and semantically unrelat ed to the target. Outside the phonology. None of the prime questions (phonological, semantic, unrelated questions for non name and proper name targets) differed from one an other in terms of word length, p s >.17 All t arget, phonological prime, semantic prime and unrelated questions can be found in the Appendix. Procedure After providing consent and completing the background questionnaire and MMSE (older adults only), partic ipants took part in the TOT elicitation task, which was presented on PC compatible computers via a program written in Visual Basic 5.0. Participants were informed that would be answering general knowledge questions
62 whose answers were famous names, uncommon words, or biographical facts about a person or event. For each question, participants were asked to indicate whether they name), did not know the answer (not familiar with th e word or name), or were having a TOT (currently unable to retrieve the word or name but certain that the answer is known and on the verge of being produced). Participants were told to respond quickly in order to reduce age differences in response times, a void excessively long testing sessions, and minimize circumstances where participants experienced a TOT but waited long enough to resolve it before moving the next question. The experimenter began the TOT elicitation task once it was clear that the partici pant was familiar with the procedure and understood the definition of a TOT. Questions were displayed in the center of the screen, approximately one third from the top of the screen. Each trial consisted of a pre target question, the target question, and a post target question, although the post target question was only presented when the participant had a TOT for the target ( Figure 3 1 for a flowchart of procedure). The target question was then re presented after the post target question in order to give the participants an opportunity to resolve their TOT. The pre target question contained the phonological prime on half of the trials and the unrelated word/name on the remaining half of trials. Two counterbalanced versions of the experiment were developed so that an equal number of participants received a given target question. Then, if the participant had a TOT for the target question, the post target question was either the phonological prime or unrel ated question, whichever one was not used as the pre target
63 question. For example, in Version 1 the Alfred Hitchcock target question was preceded by its unrelated question ( Celebrity chef Bobby Flay has a show on the Food Network that focuses on using what cooking method? ) If the participant had a TOT for the Alfred Hitchcock question, the next question would contain its phonological prime ( Which famous equation relevant to mass and energy did Albert Einstein discover? ). In Version 2, Albert Einstein would be the pre target question and Bobby Flay would be post target question. Targets alternated between non name and proper name trials, each of which was selected randomly. For each question, the experimenter recorded via keypress whether the participant sai d Known, Unknown, or TOT (pressing the letter k, u, or t, respectively). If the participant took too long to respond or was noticeably stalling, the experimenter was able to press the m key to move on to the next question. For all target questions (both th e first presentation and the second presentation if there was a TOT), the program also recorded the time interval between when the question first appeared on screen and when the experimenter pressed a key. After completing the TOT elicitation task, partici pants were given a multiple choice recognition test for all TOT questions that remained unresolved. The multiple choice test was used as a cursory measure of the correctness of TOT responses by confirming that participants could at least recognize the corr ect answer to the target question. Participants were re presented the TOT target question with four multiple choice options displayed below the question, one of which was the target. The other multiple choice options were selected as semantic associates wi th the target. For example, the alternate choices for the Alfred Hitchcock question were Cecil B. Demille Martin Scorsese and M. Night Shyamalan all film directors
64 only included in statistical analyses if participants subse quently resolved their TOTs or answered the multiple choice question correctly. Following the recognition test the participants were asked a series of questions to detect their awareness of the priming manipulation (e.g., ween the words or names in any of ). They were then asked to rate the semantic similarity between all TOT targets and their phonolo gical and unrelated primes ( Pilot study section for details on the semantic similarity ratings task). Although the semantic relationship between targets and primes had been measure d in the p ilot, the procedure was repeated in actual experiment to confirm that the participants used in the experiment perceived a close semantic relationship between the semantic prime and target, but no semantic relationship between the target and either the phonological or unrelated prime. This task was more relevant for Experiment 2 but was done in both experiments in order to keep the entire procedure consistent across experiments. Participants then completed the digit span task and vocabulary test and were debriefed and compensated (when relevant) Results For all analyses, e stimates of effect size ( r) are reported for main effects and focused contrasts. on coefficient, r is a standardized measure of the strength of an experimental effect that can be used for ANOVA when there is one degree of freedom (e.g., Field, 2005). By convention, an r value of 0.10 represents a small effect (the effect expl ains 1% o f total variance), an r value of .3 represents a medium effect (the effect explains 9% of total variance), and an r value of .5 represents a large effect (the effect explains 25% of the total variance). In analyses that were
65 conducted by participants and b y items, greater credence was given to the participant analyses due to extensive variability among items Initial Responses five categories: correct known (where they provided th e correct target name or word), incorrect known (where they said that the answer was known but provided a name or word other than the target), correct TOT (where they said TOT and subsequently resolved the TOT or correctly answered the TOT question during the recognition test), incorrect TOT (where they said TOT on the first question and then failed to resolve the TOT or get the correct answer on the recognition test, or provided an incorrect answer during the second presentation of the target question), an d unknown (where they said that the answer was unknown at the first presentation of the target question) Experimenters were instructed to skip trials where the participant was taking an excessive amount of time to respond to the question (> 10 seconds aft er reading the question), which made up a small percentage of trials (0.4% for young and 0.7% for older adults). Means for each response type as a function of age group and target type are shown in Table 3 5. Separate 2 (Age Group: Young, Older) X 2 (Targ et Type: Non Name, Name) ANOVAs were conducted on the percentage of incorrect knowns, incorrect TOTs, and unknown responses. Analyses for correct TOTs and correct Knowns are explored in depth with additional variables in the next sections. For incorrect kn own responses, only the effect of target type was significant, F (1, 103) = 157.4, MSE = .002, p < .001 r =.78 where non name questions were given more incorrect known answers relative to proper name questions. Neither the effect of age
66 group nor the age group X target type interaction was significant, p s >.151. The proportion of incorrect TOTs did not show main effects of age group or target type ps > .124, although there was a marginally significant age group x target type interaction, F (1, 103) = 2.8 6, MSE = .0002 p = .094. Older adults had more incorrect TOTs for names than non names ( p = .037 r =.2 ) while young adults had equivalent rates of incorrect TOTs for names and non names. Finally, unknown response rates showed significant main effects of age group, F (1, 103) = 45.75, MSE = .045, p < .001 r = .55 (young having more unknowns than older adults), and target type, F (1, 103) = 109.32, MSE = .008, p < .001 r =.72 (proper names having more unknown responses than non names), which were qualif ied by a significant interaction between the two variables, F (1, 103) = 12.0, MSE = .008, p < .001. This interaction reflects the fact that the higher rate of unknown responses for proper names relative to non names was more pronounced for young adults ( r =.73) than for older adults ( r =.41) and age group differences in unknown responses (young having more than older) were more pronounced for proper names ( r =.6) compared to non names ( r =.45), all p s < .001 TOT Incidence Main analyses To identify outli er items, I first calculated the mean TOT incidence rate separately for non name and proper name targets. Due to extreme item variability in TOT incidence, items that were greater than 2.5 SDs from the mean within a target category were excluded from analy ses, resulting in the loss of one non name target
67 ( kaleidoscope ) and one proper name target ( Anthony Hopkins ) 1 Prime condition in this analysis refers to whether a phonological prime or an unrelated word/name was presented in the question preceding the ta rget question. A 2 (Age Group: Young, Older) X 2 (Target Type: Non name, Name) X 2 (Prime Condition: Phonological Prime, Unrelated) X 2 (Target First Syllable Frequency: High, Low) mixed effects ANOVA was conducted by participants ( F 1 ) and items ( F 2 ) on th e mean proportion of TOT responses to target questions (# of correct TOTs/total trials). Descriptive statistics for this analysis are reported in Table 3 6. Results revealed significant main effects of age group, F 1 (1, 103) = 50.19, MSE = .023, p 1 < .001, r 1 = .57, F 2 (1, 141) = 105.2, MSE = .007, p 2 < .001, r 2 =.65, target type, F 1 (1, 103) = 43.91, MSE = .007, p 1 < .001, r 1 = .55, F 2 (1, 141) = 16.24, MSE = .012, p 2 < .001, r 2 =.3 2 and prime condition, F 1 (1, 103) = 10.28, MSE = .004, p 1 = .002, r 1 = 3 F 2 (1, 141) = 8.44, MSE = .004, p 2 =.004, r 2 =. 2 4 where more TOTs occurred for older adults, proper names, and in the unrelated condition. There was a marginal (by participants) effect of syllable frequency, F 1 (1, 103) = 3.74, MSE = .005, p 1 = .056, r 1 = .19, F 2 (1, 141) = 1.22, MSE = .012, p 2 =.272, r 2 =. 09 where more TOTs occurred for targets with HF first syllables relative to LF first syllables an inhibitory syllable frequency effect Main effects were qualified by significant two way interactions between age group and target type, F 1 (1, 103) = 14.33, MSE = .007, p 1 < .001, F 2 (1, 141) = 7.73, MSE = .007, p 2 =.006, target type and syllable frequency (by participants only) F 1 (1, 103) = 7.94, MSE = .004, p 1 = .006, F 2 (1, 141) = 2.14, MSE = .012, p 2 =.15, and marginally significant interactions between prime condition and age 1 Outliers were calculated within target type because proper name targets elicited significantly more TOTs relative to non name targets. The removal of these items did not qualitatively change the results of a nalyses
68 group, F 1 (1, 103) = 3.08, MSE = .004, p 1 = .082, F 2 (1, 141) = 2.9, MSE = .003, p 2 =.004, and prime condition and syllable frequency, F 1 (1, 103) = 2.97, MSE = .005, p 1 = .08 8, F 2 (1, 141) = 2.88, MSE = .004, p 2 =.091. All two way interactions were moderated by significant three way interactions (described below) with the exception of the prime condition X syllable frequency interaction (which did not co occur in any of the th ree way interactions). Planned comparisons revealed that priming effects were significant in reducing TOTs for targets with LF first syllables ( p 1 =.003, r 1 = .29; p 2 =.001 r 2 =. 26 ), but not for targets with HF first syllables ( p s > .373). Further, the in hibitory syllable frequency effect was significant in the phonological prime condition ( p 1 = .006, r 1 = .26, p 2 =.063 r 2 =.15 ) but not in the unrelated condition ( p s >.694). Hence, there seems to be interactivity between the influence of the prime and the number of first syllable neighbors The primary question of interest was whether age differences in TOT incidence were exacerbated for proper names relative to non names. As predicted, the age group X target type interaction revealed a disproportionate i mpairment for proper names with age: while both age groups had more TOTs for proper names, the effect was more pronounced for older adults ( ps < .001 r 1 =.56, r 2 =.33 ) than young adults ( p 1 = .029, r 1 =.23, p 2 = .041 r 2 =.17 ). Further, age differences wer e larger for names ( r 1 =.61, r 2 =.61) relative to non names ( r 1 =. 4 1, r 2 =. 41, all p s <.001). However, the magnitude of this disproportionate impairment was moderated by target first syllable frequency, as shown by a significant (by participants) age group X target type X syllable frequency interaction F 1 (1, 103) = 5.1, MSE = .004, p 1 = .026, F 2 (1, 141) = 1.87, MSE = .007, p 2 =.17 Upon further investigation, the age group X target type interaction was significant for targets
69 with HF first syllables ( p 1 <.001) but not for targets with LF first syllables ( p 1 =.119), shown in Figure 3 2. For HF first syllable targets, the age group X target type interaction is identical to the two way presented above: more pronounced effects of target type for older adults ( p 1 <.001 r 1 = .58 ) relative to young adults ( p 1 = .053 r 1 =.19 ), and more pronounced age differences for proper names ( p 1 <.001 r 1 = .59 ) relative to non names ( p 1 =.005 r 1 = .27 ). In contrast, for LF syllable targets, age differences were comparable f or proper names ( r 1 =.53) and non names ( r 1 = .45, p s <.001 ) To examine the unique role of syllable frequency in explaining age differences in TOTs for names versus non names, the age group X syllable frequency interaction was examined within each level of target type. Age differences in TOT incidence for non names was moderated by first syllable frequency, with larger older adult deficits for non names with LF first syllables ( p 1 < .00 1, r 1 =.45 ) relative to HF first syllables ( p 1 =.005, r 1 =.27) In contra st, age differences in TOT incidence for proper names was equivalent for names with HF ( r 1 =.59) and LF first syllables ( r 1 =.53), p s <.001. characteristics, the target type X sylla ble frequency interaction was examined separately for young and older adults. The target type X syllable frequency interaction was only significant for older adults ( p 1 <.001), who demonstrated opposing influences of syllable frequency for non names and pr oper names: more TOTs occurred for non names with LF first syllables versus HF first syllables ( p 1 =.049 r 1 =.19 ) and more TOTs occurred for proper names with HF first syllables versus LF first syllables ( p 1 = .015 r 1 =.24 ). In contrast, the target type X syllable frequency interaction was not significant for younger adults ( p 1 = .662), for which none of the contrasts were fully significant but
70 trended toward an inhibitory syllable frequency effect for both non names and names ( p s >.069). Overall, results from this analysis indicate that age differences in TOTs are more pronounced for proper names but only relative to non names targets with HF first names retrieval, age differences in TOTs were similar for names and non names with frequency, which exerted an opposing influence on TOTs for names and non names. There was also a marginal interaction between age group, target type and prime condition, F 1 (1, 103) = 3.38, MSE = .004, p 1 = .069, F 2 (1, 141) = 3.46, MSE = .003, p 2 =.065, displayed in Figure 3 3. The three way interaction was explained by a significant target type X prime cond ition for older adults ( p s = .025, p 2 =.062) but not for young adults ( p 1 = .89, p 2 =.076). For older adults, presentation of a phonological prime prior to the question reduced TOTs for non names relative to an unrelated word ( ps <.001 r 1 = .38, r 2 =.26 ), but priming had no effect on TOTs for proper names ( p 1 =.53, p 2 =.59). For young adults, presentation of a phonological prime had no effect on their rate of TOT incidence, for either target type, p s > .18. No other interactions involving TOT incidence as t he dependent variable were significant ( p s >.101). The disparate ranges of syllable frequencies for non name and proper name targets made it necessary to use different criteria to define HF and LF first syllables within each target type. To rule out the p ossibility that the observed effects of syllable frequency on proper name and non name TOTs were confounded by the use of different cutoffs to establish syllable frequency category, I repeated the Age Group X
71 Target Type X Prime Condition X Syllable Freque ncy analysis on the incidence of TOTs among those items that were matched on first syllable frequency (i.e., 34 pairs of non name and proper name targets with the same first syllable). The main effects and interactions reported in the previous analysis rem ained stable when only the matched stimuli were included in the ANOVA. Other proportional measures of TOT incidence A primary goal of this research was to directly test the hypothesis that aging exacerbates differences between proper name and non name ret rieval While the above analyses lend evidence in favor of selective age impairment for names, previous research has shown that the name/non name retrieval disparity may fail to emerge when the two word classes are equated for familiarity and/or name agree ment (Hanley, 2011). Young adults were less familiar with the targets (as measured by the pilot study as well as the rates of unknown responses in the current experiment), and age differences in familiarity were especially pronounced for proper names (acco rding to the rates of unknown responses reported above) with the proper name targets might reduce their opportunities for TOTs and inflate age differences in proper name TOTs. Two additional measures of TOT inci dence were calculated in order to adjust for age differences in familiarity and target type: 1) #TOTs/TOTs+ Knowns +Incorrect Knowns+ Incorrect TOTs (i.e., all trials for which a retrieval attempt was made), and 2) #TOTs/ TOTs + Knowns (all trials for whic h the participants knew the correct answer). In the first measure, unknowns were excluded in order to remove trials where participants lacked any knowledge of the answers (which disproportionately affected young adults and proper name trials) therefore ca lculating a proportional measure for which the
72 opportunities for TOTs was more similar across age groups and target types. The second measure was also used because it represents all trial s for which there was confirmation of knowledge either by producing the correct answer or getting the correct answer on the multiple choice (i.e., all trials for which a TOT was possible). B ecause the disproportionate age impairment for names was moderated by syllable frequency, 2 (Age Group) X 2 (Target Type) X 2 (Syllabl e Frequency) mixed effect ANOVAs were conducted on the se two prop ortional measures Both measures replicated the age group X target type X first syllable frequency interaction from the overall analysis (reported above): the Age Group X Target Type interact ion was significant for targets with HF first syllables ( p s <.01) but not for targets with LF first syllables ( p s >.859). For HF syllable targets, older adults had more TOTs than young adults for proper names ( p s <.001), but not for non names ( p s >.123), a nd difficulty with proper names relative to non names was more prominent among older adults ( p s <.001) than young adults ( p s <.01). While these data do not rule out the possibility that empirical evidence for a disproportionate age related impairment for n ames may be influenced by age differences in familiarity with the target stimuli, it is not the primary factor driving the age group X target type interaction reported in the current study which seems to be contingent on first syllable frequency Correct response rates To examine whether the effects of priming and syllable frequency were specific to TOTs states or could also influence other stages of lexical access, an Age Group X Target Type X Prime Condition X Syllable Frequency mixed ANOVA was conducte d on the proportion of target questions that were answered correctly on the first presentation ( # correct/total trials). Descriptive statistics for this analysis are shown in Table 3 7.
73 There were significant main effects of all variables (by participants) Older adults correctly answered more questions than young, F 1 (1, 103) = 13.2, MSE = .21, p 1 < .001, r 1 = .34 F 2 (1, 141) = 25.59, MSE = .0813, p 2 < .001 r 2 = .39 Non name questions were answered correctly more often than proper name questions, F 1 (1 103) = 39.09, MSE = .032, p 1 < .001, F 2 (1, 141) = 5.32, MSE = .151, p 2 = .022. Presenting a phonological prime resulted in higher correct answers relative to the unrelated condition, F 1 (1, 103) = 15.7, MSE = .009, p 1 < .001, r 1 =.52, F 2 (1, 141) = 11.6, MSE = .007, p 2 < .001 r 2 = .19 Finally, questions about targets with LF first syllables were answered correctly more than targets with HF first syllables (by participants only) F 1 (1, 103) = 18.29, MSE = .009, p 1 < .001, r 1 =.39, F 2 (1, 141) = 1.16, MS E = .036, p 2 = .28 r 2 =.09 There was also a significant interaction between age group and prime condition, F 1 (1, 103) = 4.54, MSE = .009, p 1 =.017, F 2 (1, 141) = 3.97, MSE = .009, p 2 = .048. Planned comparisons showed that only older adults demonstrated phonological priming effects on the proportion of known answers ( p s <.001 r 1 =.38, r 2 =.3 ); phonological primes did p 1 =.235, r 1 =.12, p 2 =.434 r 2 = .06 ) can be attributed to multiple mechanisms. One possibility is that activation spreads from the phonology of the prime to the lexical representation of the target, hence strengthening baseline activation of the nown responses), and strengthening the to phonology link (reducing the rate of TOT responses). However, phonological priming should not influence semantic level activation and therefore should as otherwise unfamiliar to the participant (affecting the rate of Unknowns), nor should it be able influence the activation level of
74 other lemmas that were semantically related to the target (affecting the rate of Incorrect Knowns). To confirm that phonolo gical priming specifically influences lexical to phonological processes but not semantic lexical processes, I compared the effect of (i.e., the age group that demonstrat ed priming effects on correct knowns). Phonological F (1, 42) = 1.88, MSE = .002, p = .177 (priming difference score = 1.1%), or incorrect known responses, F (1, 42) < 1, p = .997 (priming differ ence score = 0%). Therefore, phonological primes names) but do not influence the rates of unknown or incorrect known responses. known rates were differentially affected by target type and syllable frequency. A two way interaction between age group and syllable frequency (by participants), F 1 (1, 103) = 5.4, MSE = .009, p 1 = .022, F 2 < 1, p 2 = .57, was qualified by a three way inte raction between age group X target type X syllable frequency (by participants), F 1 (1, 103) = 11.54, MSE = .007, p 1 < .001, F 2 < 1, p 2 = .32. Unlike TOT incidence, t he target type X syllable frequency interaction was significant for both young ( p =.032) and older adults ( p = .011), although the age groups differed on when syllable frequency effects emerged. For young adults, the inhibitory effect of syllable frequency was significant for non names ( p = .01 r =.25 ) but not for names ( p = .815 r =.02). For o lder adults, the effect of syllable frequency was inhibitory for names ( p < .001 r =.43 ) but not for non names ( p =.125 r =.15 ). Inhibitory syllable frequency effects were most pronounced for older e age group X syllable
75 frequency interaction was significant for proper names, ( p < .001) but not for non names ( p = .644), as shown in Figure 3 4. L arger age differences occurred for LF syllable names ( p < .001 r =.42 ) compared to HF syllable names ( p = .017 r =.23 ), due to older In contrast, age differences in correct known rates for non names w ere unaffected by syllable frequency (both p s =.003 r s = .29 ). In sum, these analyses suggest that syllable frequency moderates age differences in correct known rates for proper names but not for non names. This contrasts with the TOT incidence analyses, where syllable frequency accounted for age differences in TOT incidence rates for non n ames but not for names. No other main effects or interactions were significant for correct known rates, p s > .322. TOT Resolution In order to examine the effect of all within subjects variables (target type, prime condition, and syllable frequency) on TO T resolution, it requires that a participant have at least one TOT in each condition (a minimum of 8 TOTs, distributed across the conditions). Because young adults had relatively low TOT rates, especially for non names, it was impossible to include all the variables of interest in a single ANOVA without losing over 80% of the young adult participants. As such, separately analyses were conducted to examine (1) the effect of phonological priming on non name and proper name TOT resolution among young and older adults, and (2) the effect of syllable frequency on primed TOT resolution in young and older adults. The same procedure was used to exclude outlier items as was used for TOT incidence, resulting in the loss of two non name items ( torpedo crucifix ). For r esolution analyses, there was greater variability and data loss among the items. As such, interactions that were only signific ant in the items analyses were not further explored
76 Phonological priming effects on TOT resolution For TOT resolution, prime con dition is defined by the type of question ( phonological prime, unrelated) presented immediately after a TOT. A 2 (Age Group) X 2 (Target Type) X 2 (Prime Condition) mixed effect ANOVA was conducted by participants and items on the proportion of TOTs that w ere successfully resolved (# function of target type and prime condition are displayed in Table 3 8. Only prime condition had a significant main effect on TOT resolution, F 1 (1, 73) = 4.03, MSE = .091, p 1 = .049, r 1 =.23, F 2 (1, 86) = 4.14, MSE = .121, p 2 = .045, r 2 =.21, where TOTs were resolved more often when the post target question contained a phonological prime (36.3%) compared to an unrelated word/name (29.3%). No other effe cts or interactions were significant in the participant analyses, p s >.116. Because the significant loss of power makes it difficult to detect higher order interactions, I compared the magnitude of priming effects separately for non name and proper name t argets, which had shown differential effects of priming on TOT incidence. Separate 2 (Age Group) X 2 (Prime Condition) ANOVAs were conducted on the mean TOT resolution rate for non name and proper name targets. Results revealed a significant priming effect for non names with higher resolution in the prime condition compared to the un related condition (priming difference score = 11.2%), F 1 (1, 77) = 4.84, MSE = .1, p 1 =.031, r 1 =.24, F 2 (1, 38) = 3.4, MSE = .15, p 2 =.07 r 2 =.29 In contrast, TOT resolution rates for proper names did not show a priming effect (priming difference score = 3.2%), Fs < .1, ps > .359. No other effects were significant, p 1 s > .188.
77 To examine this relationship the other way, I also conducted separate Age Group X Target Type ANOVAs within each level of prime condition. In the unrelated condition, there was a marginal effect of age group by participants, F 1 (1, 80) = 3.08, MSE = .115, p 1 =.083 r 1 =.19, F 2 < 1, p 2 =.334, where younger adults resolved more TOTs than older adults. No o ther effects or interactions were significant in the unrelated condition, p s >.611. In the primed condition, there was a significant main effect of target type, F 1 (1, 85) = 3.71, MSE = .077, p 1 =.057, r 1 =.2, F 2 (1, 112) = 4.73, MSE = .125, p 2 =.032, r 2 = .2, where TOTs for non names were resolved more frequently than TOTs for names, but no other effects or interactions were found, p 1 s >.586 Finally, the target type X prime condition relationship was explored within each level of age group. For young adul ts, none of the main effects or interactions were significant, p s > .205, suggesting no phonological priming for either target type For older adults, there was a main effect of prime condition, F 1 (1, 40) = 4.72, MSE = .086, p 1 =.036, r 1 =.32, F 2 (1, 124) = 5.63, MSE = .094, p 2 =.019 r 2 =.21 reflecting the increase in resolution following a prime word versus an unrelated word, and to the same degree for both names and non names. None of the other effects were significant, p 1 s > .1. Effects of syllable fr equency on primed TOT resolution To investigate how syllable frequency might interface with phonological priming during resolution of proper name and non name TOTs, a 2 (Age Group) X 2 (Target Type) X 2 (Syllable Frequency) ANOVA was conducted on mean TOT resolution rates in the primed condition. Syllable frequency effects were not examined in the unrelated condition because there were too few cases (only 32% of participants could be included in the analysis). Operating on the assumption that primes influen ce TOT resolution via
78 the primed condition relative to the unrelated condition. Results revealed a marginally significant interaction (by participants only) betw een target type and syllable frequency F 1 (1, 47) = 3.22, MSE = .085, p 1 = .079, F 2 < 1, p 2 = .85 (shown in Figure 3 5 collapsed across Age Group). Focused coontrast revealed than names with HF first syllables were resolved less often than names with LF f irst syllables ( p 1 = .06 r 1 =.27 ). There was no effect of syllable frequency on non name TOT resolution ( p 1 = .69 r 2 =.06 ), Examined the other way, non names were resolved more often than proper name targets when the targets had HF first syllables ( p 1 < .0 01 r 1 =.47 ); proper names with LF first syllables were resolved as often as non names with LF first syllables ( p 1 =.317 r 2 = .14 ). No other effects or interactions involving syllable frequency were significant, p s >.442. Partial name retrievals To be cod ed as a correct response or a correctly resolved TOT, we required that added level of difficulty of retrieving two separate lexical entries (for the first and last n ame) is the source of the disproportionate impairment for names, as opposed to differences in the representation of names or the processes engaged during their retrieval or last name was correctly provided during the first presentation of the target question, the second presentation of the target question, or when participants were asked to give any available information about the TOT name following the second presentation of the target questions. Overall, participants very rarely had access to one of the names but not the other (1.8% of proper name trials for young, 2.5% of proper name trials for older). When these partial answers were coded as correct responses, it did not c hange
79 the pattern of proper name versus non name retrieval described i n the previous analyses. Discussion Results from this experiment lend further empirical support in favor a disproportionate age related impairment for proper names, but more importantly they clarify the conditions in which selective deficits occur. Fundamentally, age differences in TOTs for proper names and non names were differentially affected by two constructs that were designed to tap into phonological level processes during TOT sta tes: first syllable frequency and phonological priming. Consistent with previous research (Burke et al., 1991; Cohen & Faulkner, 1986; Evrard, 2002; James, 2006; but see Maylor, 1995; 1997), age differences in TOT incidence were exacerbated for proper name s However, the unique finding of the present experiment was that this particular age impairment for names only occurred when the targets contained HF first syllables. Age differences were comparable for names and non names when they began with a LF first syllable, suggesting that phonological frequency is a critical determinant of older (but not young) TOTs. The effectiveness of phonological priming also differed as a function of age group and target type. P honological primes reduced TOTs, increase d correct knowns, and bolstered TOT resolution, but only for older adults and only for non names. The specificity of priming effects observed in this experiment suggests that priming is only effective when it is able to activate the precise link that trigg ered the TOT, which may be more difficult to access for proper names and young adults. Taken together, these results imply the type of phonological retrieval failures that result in a TOT may differ for non names and proper names and may also differ for yo ung and
80 older adults, providing insight into how these word classes are represented in the phonological lexicon and how they are processed throughout the lifespan. A primary first goal of this project was to assess whether there is a particular age relat ed deficit during the retrieval of proper name s relative to other types of words, which has been challenged by a few studies that reported comparable age deficits during the retrieval of proper names and other types of information (e.g., Maylor, 199 7 ). As names, but only when proper names are being compared to non names with HF first syllables. Age differences in TOT incidence for non names with LF first syllables were si milar to that of proper n ames, suggesting that different factors may account for age related declines in the retrieval of names versus non names. To my knowledge, this is the first study to report an opposing influence of the same variable (first syllable frequency) on the processing of proper names and non names. Older adults had more TOTs for non names with LF first syllables relative to HF first syllables, replicating previous research (Farrell & Abrams, 2011) but showed the opposite effect for names, h aving more TOTs for names with HF first syllables relative to LF first syllables. Older phonologically related names which may be caused by confusability or competition among alte rnative name options (is it Jennifer, Jessica, Jenna?) an interference that does not occur for non names Interference from similar sounding first names may also explain why phonological priming of proper names has been elusive, at least when the primes contain only a single syllable (e.g., White et al., in press) At a structural level, there is an additional
81 level of lexical phonological selection that is demanded by names but not by non names which enhances the opportunity for competition: first name selection Once a presented in the question (e.g., the Julie Andrews node), the speaker then has to then select the first name (Julie) and last name (Andrews) label that cor responds to that person, followed by the phonological subcomponents contained within the names. Theoretically, primes reduce TOTs and/or promote TOT resolution by activating the weakened link that caused the TOT in the first place, i.e., the link between t lemma and its first syllable. However, if the lemma to first syllable link was NOT the one that caused the original transmission failure, then a first syllable prime would not be helpful and may in fact introduce competitors by spreading activa tion to other first names The lack of phonological priming of proper names implies that proper name syllable but may be caused by an inability to access the first or last name. Phonological primes did facilitate retrieval of non names, but only for older names are specifically driven by weakened connections to LF first syllables. As such, the presentation of a word with the same first syllable is able to activate this weakened connection, thereby preventing TOTs and bolstering TOT resolution. Conversely, incidence and did not signific ant ly al though resolution was qualitatively higher in the primed, M = 44.1, versus unprimed condition, M = 34.9). The ~ 10%
82 (significant) p riming effect ( 12%), so the lack of significance may represent the young adults showed no evidence of priming for TOT incidence, which cannot be attributed to inadequate statistical power This is the first study to report significant phonological priming effects for older adults but not for young at least within the context of TOTs. Methodologically, this study is unique in that it is the first to attempt priming of TOT incidence using onl y a single syllable (and embedded within a questi on), so it may have uncovered a unique age related discrepancy in the amount of phonology that is necessary to prevent TOTs. Overall the results of this experiment suggest that spreading activation from a s ingle syllable does not provide enough information to prevent TOTs from occurring but may help young participants resolve their TOTs once lexical selection has occurred (as has been shown in previous studies) In sum, results from Experime nt 1 suggest that age differences in TOTs for non names and names stem from different sources of failure during lexical access. Older names are specific to phonological retrieval failures, as evidenced by facilitatory syllable frequenc y effects and significant priming from first syllable neighbors. TOTs for proper names may arise from breakdowns at multiple locations in the lexical phonological system, as evidenced by inhibitory syllable frequency effects (competition from similar sound ing names) and null phonological priming effects (due to indirect links between the lemma associated with the target and its first syllable). These findings expand on and clarify existing theoretical accounts of the specialness of proper names by adding gr eater specificity to where and how proper names diverge from other word classes. In Experiment 2, I attempted to gain insight into structural differences
83 between names and non names at the level of semantic activation and lexical selection. In the General Discussion, the implications of these results will be discussed in greater detail in conjunction with findings from Experiment 2 to develop a more comprehensive model of how proper names are represented and accessed.
84 Table 3 1. Descriptive characteristic s for young and older participants in Experiment 1 Age Group Young Older M SD M SD Age* 19.4 1.6 68.2 4.7 Health Rating (out of 10) 8.3 1.2 7.9 2.3 Education (in years)* 13.3 1.2 16.4 2.8 Vocabulary (out of 25)* 14.2 4.3 18.6 5.4 Forward Digit S pan* 7.5 1.2 6.4 2.1 Backward Digit Span 5.1 1.2 5.0 1.2 MMSE --28.5 1.7 Note: indicates significant age group differences (p <.001).
85 Table 3 2. S ample non name targets with phonological, semantic, and unrelated primes Sample Non Names with HF First Syllables ( M = 2023.4) Target Phonological Prime Semantic Prime Unrelated Acronym Accurate Initials Surprising Arson Arbitrary Pyromania Ridiculous Marsupial Marginal Kangaroo Telephone Obituary Obey Eulogy Ignore Sample Non Names wi th LF First Syllables ( M = 230.7) Target Phonological Prime Semantic Prime Unrelated Barnacle Barter Oyster Cookie Beaker Beaten Flask Pleasant Embryo Empathize Fetus Revisit Haiku Hygienic Limerick Cabinet Table 3 3. Sample proper name targets wi th phonol ogical, semantic, and unrelated primes Sample Non Names with HF First Syllables ( M = 1722.8) Target Phonological Prime Semantic Prime Unrelated Alexander Hamilton Alec Baldwin John Adams Bob Dylan Arthur Miller Arnold Palmer Neil Simon Ryan Rey nolds Leonardo da Vinci Lee Harvey Oswald Michelangelo Paul Michael Glaser Monica Lewinsky Molly Ringwald Paula Jones Kelly Clarkson Sample Non Names with LF First Syllables ( M = 150.8) Target Phonological Prime Semantic Prime Unrelated Julia Ch ild Judy Blume Martha Stewart Katharine Hepburn Lucille Ball Louisa May Alcott Carol Burnette Erin Andrews Rosa Parks Coretta Scott King Marie Antoinette Rudy Giuliani Rupert Murdoch Michael Bloomberg Conrad Hilton
86 Table 3 4. Mea n word length of each type of target and prime question Target Type Target Question Phonological Prime Question Semantic Prime Question Unrelated Question Non Name 23.5 21.7 21.7 21.0 Proper Name 24.7 20.8 22.2 21.1 Table 3 5. Percentage of each respo nse type as a function of age group and target type Age Group Young Old er Non Names Correct Known 41.9 52.3 Incorrect Known 15.3 14.6 Correct TOT 5.9 11.6 Incorrect TOT 1.9 1.6 Unknown 34.5 18.8 Proper Names Correct Known 32.3 45.0 Inco rrect Known 6.0 7.2 Correct TOT 7.3 17.6 Incorrect TOT 1.8 2.3 Unknown 52.4 27.8
87 Table 3 type, prime condition, and first syllable frequency. Age Group Young Old er M SE M SE Non Names Primed HF First Syllable 6.2 1.0 9.4 1.2 LF First Syllable 4.5 1.0 8.8 1.1 Unrelated HF First Syllable 6.3 1.0 10.8 1.1 LF First Syllable 5.6 1.1 15.3 1.3 Proper Names Primed HF First Syllable 7.9 1 .1 18.0 1.3 LF First Syllable 5.9 1.2 15.3 1.4 Unrelated HF First Syllable 8.4 1.1 18.8 1.4 LF First Syllable 6.9 1.1 16 1.3
88 Table 3 function of target type and prime conditio n Age Group Young Old er M SE M SE Non Names Primed HF First Syllable 40.3 2.6 52.0 3.1 LF First Syllable 43.0 2.5 56.2 3.0 Unrelated HF First Syllable 39.0 2.5 50.4 3.0 LF First Syllable 42.2 2.4 50.4 2.8 Proper Names Primed HF First Syllable 33.2 2.3 43.9 2.8 LF First Syllable 33.1 2.4 50.6 2.9 Unrelated HF First Syllable 32.4 2.5 39.2 3.0 LF First Syllable 32.0 2.2 45.9 2.7
89 Table 3 on of target type and prime condition. Age Group Young Older M SE M SE Non Names Primed 44.1 5.7 37.9 5.2 Unrelated 34.9 6.1 25.7 5.6 Proper Names Primed 30.9 4.3 32.3 3.9 Unrelated 31.8 4.8 24.7 4.4
90 Give Answer: K, U, or TOT Give Answer: K, U, or TOT Give Answer: K, U, or TOT Prime before Target Question Prime after Target Question If K: Produce Name Next Trial Which English filmmaker was the master of suspense, making films such as Psycho and Vertigo? Which famous equation relevant to mass and energy did Albert Einstein discover? Give Answer: K, U or TOT Which English filmmaker was the master of suspense, making films such as Psycho and Vertigo? Unknown Celebrity chef Bobby Flay has a show on the Food Network that focuses on using what cooking method? Give Answer: K, U, or TOT Celebrity chef Bobby Flay has a sh ow on the Food Network that focuses on using what cooking method? Which famous equation relevant to mass and energy did Albert Einstein discover? Figure 3 1. Trial structure for Experiment 1
91 Figure 3 2. Age differences in TOT incidence as a function of target type and first syllable frequency. 0 2 4 6 8 10 12 14 16 18 20 Non-Name Proper Name TOT Incidence (in %) TOT Incidence for HF Syllable Targets Young Older 0 2 4 6 8 10 12 14 16 18 20 Non-Name Proper Name TOT Incidence (in %) TOT Incidence for LF Syllable Targets Young Older
92 Figure 3 3. Age differences in phonological priming of TOT incidence. 0 2 4 6 8 10 12 14 16 18 20 Non-Name Proper Name TOT Incidence (in %) Target Type Young Adults' TOT Incidence Phonological Prime Unrelated 0 2 4 6 8 10 12 14 16 18 20 Non-Name Proper Name TOT Incidence (in %) Target Type Older Adults' TOT Incidence Phonological Prime Unrelated
93 Figure 3 4. Age differences in correc t known responses as a function of target type and first syllable frequency. 0 10 20 30 40 50 60 HF First Syllable LF First Syllable Correct Knowns (in %) First Syllable Frequency Category Correct Knowns for Non Names Young Older 0 10 20 30 40 50 60 HF First Syllable LF First Syllable Correct Knowns (in%) First Syllable Frequency Category Correct Knowns for Proper Names Young Older
94 Figure 3 5. Effects of syllable frequency on primed TOT resolution of non name and proper name targets (collapsed across age group).
95 CHAPTER 4 E XPERIMENT 2: SEMANTI C INFLUENCES ON NON NAME AND PROPER NAME TOTS In Experiment 2 a semantic prime (or competitor) was introduced to the TOT task, a manipulation that should theoretically influence lexical activation but not phonological level processes. Previous research is equivocal regarding the influence of semantic primes on successful word retrieval, and especially TOTs. While some studies report null effects of semantic priming on TOT incidence for proper names (Cross & Burke, 2004) and non names (M eyer & Bock, 1992, Experiment 1), interference has been reported in some contexts (Meyer & Bock, 1992, Experiment 2). In other methodologies, picture naming of objects is generally slowed by the presence of a semantically related distractor word, an interf erence effect ascribed to competition from the distractor at lexical selection (e.g., Schriefers et al., 1990). In contrast, pictures of famous faces are named more accurately when they are preceded by a picture of a semantically related person relative to an unrelated person (Vitkovitch et al., 2006), suggesting that the selection of proper names does not occur in competition, as does lemma selection of non proper names. While semantic priming effects have never been directly compared for proper names and non names, there are theoretical reasons to suspect that the presence of a semantic prime may exert differential influences on the two word types during lexical selection as well as divergent effects on TOT incidence and the proportion of correct responses Of critical interest is how semantic primes will bolster (or inhibit) correct responding relative to the phonological priming effects found in Experiment 1. Unlike phonological priming which is designed to offset age declines in phonological retrieval,
96 s emantic priming is not a compensatory tool for older adults, who typically have intact or superior semantic knowledge relative to younger adults. Due to the vastness and richness of semantic knowledge, they could potentially benefit more from semantic prim ing than young adults, as they have more converging connections from the semantic level. Finally, whether semantic priming interfaces with syllable frequency will provide an interesting test of the limits of spreading activation and parallel activity withi n the production network. If more or less semantic priming occurs as a function of syllable frequency, it would challenge the notion of fully distinct semantic, lexical, and phonological systems. Specific Aims Specific Aim 4 To examine the interaction betw een semantic and phonological factors on TOT incidence and correct retrieval of proper names in young and older adults. As discussed in the introduction, the disproportionate deficit for name retrieval in older age may occur as a function of semantic, lex ical, and phonological factors. As Experiment 1 was designed to tap into the phonological characteristics of names and non names and their interactions with lexical competition, Experiment 2 attempted to illustrate differences between the two word classes at earlier stages in the production process, semantic activation through lexical selection. Semantic priming was used as an indirect indicator of semantic organization. There is evidence to support predictions that semantic priming of word retrieval may be enhanced (e.g., Burgess & Conley, 1998 ) or reduced (e.g., Bredart et al.,1995; Schweinberger, Burton, & Kelly, 2001) in proper names relative to non names. According to Burgess and Conley (199 8 ), proper names form a close semantic network specific to indi vidual people. As a result,
97 semantic activation may spread more efficiently between nodes representing semantically related persons, resulting in more pronounced semantic priming effects for proper names versus non names, which have more diffusely distribu ted semantic representations. In contrast, proper name retrieval may show reduced semantic priming if biographical information about people is stored in a separate module than names (e.g., Bredart et al., 1995; Schweinberger et al., 2001) and is not automa tically accessed during name retrieval. In support of the latter idea, Schweinberger failed to find semantic context priming when participants were asked to make syllabic judgments about proper names. Importantly, semantic primes were expected to influence correct responses rates but not TOTs, the opposite pattern than was projected for phonological primes Older adults were predicted to show more pronounced semantic priming effects relative to young due to their possession of a larger, more complex semanti c network. An interaction between semantic priming and phonological frequency was not ex pected but would imply greater parallel activity among levels in the network than is assumed by existing models of speech production and face recognition. Specific Aim 5 To check for semantic priming of TOT resolution among proper names and non names, an effect that would challenge the purely phonological locus of TOT states To my knowledge, no previous study has assessed the influence of a semantic prime on the retr ieval of a target word once already in a TOT state. Presenting a semantic prime afte r a TOT has occurred will provide an interesting test of theoretical explanations of TOTs. If current assumptions hold true, the presentation of a semantic prime should not be able to activate the blocked lemma to phonology connection that
98 phonological primes influenced in Experiment 1. As such, semantic prim es were also not expected to interact with the effects of age, target type, or target first syllable frequency on TOT resolution. However, if semantic priming effects do emerge, they would imply that TOTs can result from blockages at multiple levels in the production system, such as the link between the semantic system and the lexical level. Other studies have referred to unable to activate the correct lexical representation following semantic activation (e.g., Gollan & Brown, 2006; Hanley, 2011). Stage 1 failures were assumed to be represented feeling of pending retrieval associated with TOTs can occur at the semantic to lemma link in addition to the lemma to phonology link. Notwithstanding, the consistency of phonologic al priming effects across TOT studies (lending support to the phonological locus of TOT states) leads to the hypothesis that semantic primes will not influence the rate of TOT resolution or interact with phonological level variables during resolution. Met hod Participants The same inclusion criteria were used in Experiment 2: native English speakers, normal or corrected to normal vision and hearing, no diagnosis of a learning disability or cognitive impairment, and a score of 25 or above on the MMSE (for older adults). Participants were recruited through the same methods described in Experiment 1. Young adults were either given credit toward course requirements or $10 compensation, and all older adults were paid $8/hour for their time Data from one young adult were removed from analysis due to English being the second language.
99 Data from three older adults were removed from analysis for scoring below 25 on the MMSE. The removal of these participants resulted in 59 young adults ( M = 19.4, SD = 1.6, range = 18 25 years) of which 40% male and 60% were female, and 40 older adults ( M = 69, SD = 4.4, range = 60 75) of which 35% were male and 65% were female Descriptive statistics on background characteristics (years of education, health rating) and cognitive m easures (forward digit span, backward digit span, vocabulary, MMSE) are shown in Table 4 1. Independent samples t tests revealed that older adults had more years of education and higher vocabularies than young adults ( p s <.001), but the age groups did not differ on the other dimensions ( p s > .356). Materials The same 76 proper names and 76 non name nouns from Experiment 1 were used as targets in this experiment (Experiment 1 method for details). Each target was also paired with a semantic prime word/name a nd an unrelated word/name, the same unrelated items that were used in Experiment 1. Semantic non name primes were also nouns and were close semantic associates with the target (e.g., flask for the target beaker ), a subordinate/superordinate member of the s ame semantic category as the target (e.g., kangaroo for marsupial ), closely related in meaning to the target (e.g., metaphor for simile ), or shared a common semantic concept (e.g., eulogy for obituary ). Semantic proper name primes came from the same occupa tional or conceptual category as the target, which included actors, musicians, authors, politicians, athletes, to Table 3 2 for examples). In order to enhance the seman tic overlap between targets and primes, efforts were made to find the closest semantic alternative to a given target
100 by selecting an individual from the same specific genre of work or artistry, i.e., the same literary genre ( Mark Twain for Ernest Hemingway ), played the same sport or position ( Ryan Lochte for Michael Phelps ), held the same office ( Hillary Clinton for Condaleeza Rice ) and/or was involved in a common project or film ( Tommy Lee Jones for Harrison Ford ). Target names that are fictional characte rs were paired with semantic primes that are also fictional characters within the same modality (e.g., Olive Oyl for Betty Boop ). If primes were mostly individuals with the same gender and ethnicity as the target, with the exception of two targets ( Condaleeza Rice Mahatma Ghandi ). All unrelated words and names shared no phonological or semantic features with the target. The same targ et questions and unrelated questions described in Experiment 1 were used in this study. The semantic prime questions were designed to activate semantic commonalities between the target and prime by making specific mention of shared semantic features. For e xample, the semantic prime question for the target Harrison Ford included the shared film The Fugitive One of Tommy Lee Jones' most notable roles was a federal marshal in which movie about an on the run doctor wrongfully accused of mu The semantic prime questions were matched on word length with both the unrelated questions and the phonological prime questions ( Table 3 4 ). Procedure The procedure was identical to Experiment 1 except that semantic primes were used instead of phon ological primes during the TOT elicitation task. During the TOT
101 elicitation task, target questions were preceded by either the semantic prime question or the unrelated question. If the participant had a TOT for the target, the post target question was eith er the semantic prime or unrelated question, whichever one had not been used prior to the target (Figure 4 1 for an example of the procedure with a non name target and semantic prime). Participants were randomly assigned to one of two versions of the exper iment to ensure that equal numbers of participants received the semantic prime and unrelated question as the pre target question for a given target. As with the first experiment, the TOT elicitation task was followed by a multiple choice recognition test, and then a questionnaire to assess their awareness of the priming manipulation and any explicit strategies they may have invoked. They then rated the semantic similarity between each TOT target and its semantic prime and unrelated prime on a 3 point scale (1= completely unrelated, 2= somewhat related, 3= strongly related; Pilot section for details). Following the computer program the participants completed the digit span and vocabulary tests and were then debriefed and given compensation. Results Initial Responses Experimenters were again instructed to skip trials where the participant took too long to respond (> 10 seconds after reading the question), accounting for 0.6% of trials for young adults and 1.3% of trials for older adults. Initial responses we re classified into the same 5 categories described in Experiment 1: correct known, incorrect known, correct TOT, incorrect TOT, and unknown. Means for each response type as a function of age group and target type are shown in Table 4 2. Two (Age Group: You ng, Older) X
102 2 (Target Type: Non Name, Name) ANOVAs were conducted on the percentage of incorrect knowns, incorrect TOTs, and unknowns (as the proportions of TOTs and correct knowns are explored in depth with other variables). For incorrect known responses the main effect of target type, F (1, 96) = 158.48, MSE = .002, p < .001, r =.79, and marginal effect of age group, F (1, 96) = 3.34, MSE = .004, p = .071, r =.18, were moderated by a significant age group X target type interaction, F (1, 96) = 5.86, MSE = .002, p = .017. Planned comparisons revealed that older adults had more incorrect knowns for proper name targets ( p <.001 r =.39 ) but did not differ from young adults for non name targets ( p = .951). The proportion of incorrect TOTs did not show any ma in effects or interactions, p s > 2.66. Finally, unknown response rates showed significant main effects of target type, F (1, 96) = 125.79, MSE = .008, p < .001, r =.75, and age group, F (1, 96) = 81.62, MSE = .038, p < .001, r =.68, which were qualified by a significant interaction between the two variables F (1, 96) = 11.62, MSE = .008, p < .001. This interaction reflects the fact that the higher rate of unknown responses for proper names relative to non names was more pronounced for young adults ( r =.76) t han for older adults ( r =.46, both p s < .001), and age differences (young having more unknowns than old er ) were larger for proper names ( r =.67) relative to non names ( r =.6, both p s <.001). Semantic Similarity Ratings bout the semantic similarity between targets and primes and to see whether this relationship differed as a function of age, mean similarity ratings for all TOT targets and their semantic and unrelated primes were compiled Items for which the participant r esponded that they were unfamiliar with either
103 the prime or target (and could not assess their similarity) were excluded (19.6% for young, 5.5% for older) A 2 (Age Group) X 2 (Prime Type: Semantic Prime, Unrelated) ANOVA revealed that semantic primes were rated as more similar to their targets than the unrelated primes, F 1 (1, 78) = 3107.45, MSE = .058, p 1 <.001, an important check on the semantic priming manipulation that did not differ as a function of age group F 1 (1, 78) = 2.03, MSE = .058, p 1 =.157. TOT Incidence The same procedure was used to exclude outlier items as was done in E xperiment 1, i.e., items that were +/ 2.5 SDs from the mean TOT incidence rate for a given target type. This resulted in the loss of one non name target ( acronym ), which w as excluded from all TOT incidence analyses. Main analyses A 2 (Age Group: Young, Older) X 2 (Target Type: Non Name, Proper Name) X 2 (Prime Condition: Semantic Prime, Unrelated) X 2 (First Syllable Frequency: HF, LF) mixed effect ANOVA was conducted on mean TOT incidence rate by participants ( F 1 ) and items ( F 2 ). Means from this analysis are shown in Table 4 3. Unlike Experiment 1, prime condition had no influence on TOT incidence, F 1 < 1, p 1 = .337, F 2 (1, 142) = 1.54, MSE = .004, p 2 =.217. Independent of prime condition, the results from Experiment 1 were largely replicated in Experiment 2. There were significant main effects of target type, F 1 (1, 96) = 51.48, MSE = .007, p 1 < .001, r 1 =.59, F 2 (1, 142) = 20.39, MSE = .013, p 2 <.001, r 2 = .35, and age gr oup, F 1 (1, 96) = 16.35, MSE = .034, p 1 < .001, r 1 = .38, F 2 (1, 142) = 45.56, MSE = .009, p 2 <.001, r 2 =.49, where more TOTs occurred for proper names and older adults. There were also significant two way interactions between age
104 group and target type, F 1 (1, 96) = 11.44, MSE = .007, p 1 < .001, F 2 (1, 142) = 5.88, MSE = .009, p 2 =.017, and target type and first syllable frequency (by participants), F 1 (1, 96) = 5.19, MSE = .005, p 1 = .025, F 2 (1, 142) = 1.42, MSE = .013, p 2 =.235. As in Experiment 1, there was a significant age group X target type X syllable frequency interaction (by participants), F 1 (1, 96) = 5.93, MSE = .005, p 1 = .017, F 2 (1, 142) = 1.9, MSE = .009, p 2 =.17, depicted in Figure 4 2. Age differences in TOT incidence were contingent on tar get type and syllable frequency category. The age group X target type interaction was significant for items with HF first syllables ( p 1 <.001) but not LF first syllables ( p 1 =.301), replicating Experiment 1. When targets possessed HF first syllables, older adults had more TOTs relative to young for proper names ( p 1 <.001 r 1 =.38 ), but equivalent TOT rates as young for non names ( p 1 =.388 r 1 =.09 ). This pattern deviated slightly from Experiment 1, where older adults had more TOTs than young in all conditions but the size of age differences were smaller for non names with HF first syllables compared to proper names with HF first syllables. In contrast, the magnitude of age differences were similar for names ( r 1 =.38) and non names ( r 1 =.4) when they possessed a LF first syllable ( p s < .001), which occurred across both experiments. For non names, age differences in TOT incidence was dependent on syllable frequency category, where older adults only had more TOTs than young for non names with LF first syllables ( p 1 < .00 1, r 1 =.4 ); young and older adults had similar TOT rates for non names with HF first syllables. In contrast, for names, age differences in TOT incidence did not differ as a function of syllable frequency ( p s <.001), with older adults having more TOTs than young adults for both LF and HF syllable names (consistent
105 with Experiment 1). The target type X syllable frequency interaction was only significant for older adults ( p 1 =.003), because they exhibited opposing effects of syllable frequency as a functi on of target type: more TOTs for non names with LF first syllables relative to HF first syllables ( p 1 =.039 r 1 =.21 ), and more TOTs for names with HF first syllables relative to LF first syllables ( p 1 =.056 r 1 =.19 ). Young adults trended toward an inhibito ry syllable frequency effect for both target types ( p s >.06 2 ). No other effects or interactions were significant, p s >.112. Correct response rates As expected, semantic priming did not influence the occurrence of TOT states, which is linked more strongly semantic priming might affect other aspects of lexical access, the Age Group X Target Type X Prime Condition X Syllable Frequency ANOVA was also conducted on the proportion of correctly answered target qu estions (# Correct Knowns/Total Trials). Means from this analysis are shown in Table 4 4. There were significant main effects of age group, F 1 (1, 96) = 33.89, MSE = .165, p 1 < .001, r 1 =.51, F 2 (1, 14 2 ) = 6 3 01 MSE = .07, p 2 <.001, r 2 =.55, target type, F 1 (1, 96) = 34.25, MSE = .037, p 1 < .001, r 1 =.51, F 2 (1, 14 2 ) = 8.11 MSE = 143 p 2 =.00 5 r 2 =.23, and syllable frequency (by participants) F 1 (1, 96) = 5.57, MSE = .06, p 1 = .02, r 1 = .24 F 2 <1 p 2 = 497 where more correct answers occurred for older adults, non names, and LF first syllable targets. Although the main effect of prime condition was not significant, F 1 (1, 96) = 1.78, MSE = .006, p 1 = .185, F 2 (1, 14 2 ) = 1. 99 MSE = .009, p 2 =. 161 there was a significant interaction between prime condit ion and age group, F 1 (1, 96) = 12.48, MSE = .006, p 1 < .001, F 2 (1, 149 2 ) = 8.91 MSE = .00 6 p 2 =.00 3 Planned comparisons
106 within the age group X prime condition interaction revealed that older adults had a significant reduction in correct response rates following semantic primes ( p 1 = .002, r 1 =. 3 p 2 = .01 r 2 =.21 ); younger adults trended in the same direction though the effect was not significant ( p 1 =.1 r 1 = .17, p 2 =.338 r 2 =.08 ). These effects were qualified by a significant age group X target t ype X syllable frequency interaction (by participants) F 1 (1, 96) = 7.28, MSE = .009, p 1 = .008, F 2 <1 p 2 =.36 and a marginally significant four way interaction F 1 (1, 96) = 3.15, MSE = .01, p 1 = .079, F 2 (1, 14 2 ) = 5.48 MSE = .00 6 7, p 2 =.0 21 Because the primary question of interest was how the semantic primes would influence correct response rates, I examined the effect of prime condition within each target/syllable frequency combination as a function of age. The target type X prime condition X syllab le frequency interaction was significant for older adults, p 1 =.053, p 2 =.073, rate was reduced by semantic primes, but only for non names with HF first syllables ( p 1 =.006 r 1 =.28, p 2 =.011 r 2 = .21 ). There was no influence of semantic primes for non names with LF first syllables ( p 1 = .245 p 2 =.11 ), or for proper names from either syllable frequency category ( p 1 s > .139). In contrast, t he target type X prime condition X syllable frequency interaction was not signi ficant for young adults, p 1 = .680, p 2 =.231, response rate in any target type/ syllable frequency condition ( p s >. 19 ). I also examined the age group X target type X syllabl e frequency interaction separately in the semantic prime and unrelated conditions. The three way interaction was only significant in the unrelated condition (by participants) p 1 =.003 p 2 =.11 As in Experiment 1 the target X syllable frequency interacti on was significant for older adults
107 ( p 1 =.009) reflecting an inhibitory syllable frequency effect for names ( p 1 =.004 r 1 =.29 ) but not for non names ( p 1 =.471 ). Unlike Experiment 1, the target X syllable frequency interaction was not significant for young ( p 1 =.119), but the comparisons trended in the same direction, with young showing a marginal inhibitory syllable fre quency effect for non names, p 1 = .075, r 1 =.18, and no effect of syllable frequency for names, p 1 = .743. As in Experiment 1, the signific ant age X syllable frequency interaction for proper names ( p =.014), revealed that age differences were larger for proper names with LF first syllables ( r 1 = .56) compared to proper names with HF first syllables ( r 1 =.39. both p s <.001). The age group X syll able frequency interaction was also marginal for non names ( p = .091) where age differences were larger for non names with HF ( r 1 =.53) first syllables compared to non names with LF first syllables ( r 1 =.39, both p s <.001) Overall, patterns in the unrelat ed condition were similar to Experiment 1, with the exception of syllable frequency moderating age differences in both target types here (though marginally for non names ), and only proper names in Experiment 1. The semantic prime altered these patterns by name targets with HF first syllables (the condition where age differences were most pronounced in the unrelated condition and Experiment 1), therefore making age differences more similar across conditions ( Figu re 4 3 ) As such, the three way interaction was not significant in the semantic condition p 1 = .532, p 2 = .843. No other effects or interactions involving correct known rates were significant, p >.136. Theoretically, the semantic prime would influence cor rect response rates by interfering with lexical selection of the target. There are two methods by which lexical selection of the target would be hindered by the prime: (1) by activating the semantics of
108 several associated words and concepts, it prevents or delays the activation of the (2) by bolstering activation of alternate lemma(s) (resulting in fewer correct knowns and Older, but not youn g, adults exhibited a reduction in correct knowns following a semantic prime, so furthe r analysis on the older adults can help to dissociate these two possibilities. A target type X prime condition ANOVA was conducted (by participants) portion of unknown answers (to test the first hypothesis) and on proportion of incorrect known responses (to test the second hypothesis) effect of target type, F (1, 39) = 31.87, MSE = .014, p <.001, r = .67, with fewer unknowns for non names ( M = 17.4%) relative to proper names ( M = 27.9%). Neither the effect of prime condition, F < 1, p = .774, nor the prime condition X target type interaction, F (1, 39) = 1.22, MSE = .004, p =.28, had a significant impact on older However, in support of the alternative lemma hypothesis, the semantic prime increased the rate of incorrect known responses ( M = 13.4%) relativ e to the unrelated condition ( M = 11.6%), F (1, 39) = 4.4, MSE = .003, p = .043 r = .32 known responses was also affected by target type, F (1, 39) = 64.78, MSE = .012, p <.001, r = .79, with more incorrect knowns for non names ( M = 16.2%) than names ( M = 8.9%). The prime condition X target type interaction was not significant, F < 1 p = .801, suggesting equivalent semantic priming effects on incorrect knowns for names and non names among older adults.
109 One relevant questi on is whether the increase in incorrect knowns following semantic priming resulted from the prime word itself being produced in lieu of the target. Overall, if an incorrect response was given to a target question in the semantic prime condition, 31% of the time it was the semantic prime itself. In contrast, the semantic prime was never offered as an incorrect response in the unrelated condition. To further examine the likelihood of producing the semantic prime instead of the target, an Age Group X Target Ty pe ANOVA was conducted on the proportion of times a semantic prime was given as the answer in the semantic prime condition (# of times semantic time produced/total semantic prime trials). Non name semantic primes ( M = 5.3%) were more likely to be given as the answer relative to proper name semantic primes ( M = 1.9%), F (1, 96) = 64.67, MSE = .001, p <.001 r = 63 An interaction between age group and target type, F (1, 96) = 4.49, MSE = .001, p =.037, revealed that older adults ( M = 2.4%) were more likely t han young adults ( M = 1.4%) to give the semantic prime as the answer for proper name questions ( p =.053 r =.2 ), but that the age groups were equally likely to produce the semantic primes for non name questions ( M Older = 4.9%, M Young = 5.7%, p = .307). TO T Resolution Targets whose mean TOT resolution rate was more than 2.5 SDs from the target type mean were excluded from analysis, which included three non names ( diameter, lentil, allegory ) and one proper name ( Carmen Electra ). As with Experiment 1, extreme data loss prevented the ability to look at all variables of interest within a single analysis. Because semantic primes were not expected to interface with syllable frequency (and did not interact with syllable frequency on TOT incidence), I examined the A ge Group X
110 Target Type X Prime Condition relationship (collapsed across syllable frequency category, shown in Table 4 5 ), and Age Group X Target Type X Syllable Frequency interaction (collapsed across prime condition). Because the items were associated wit h high variability in TOT resolution, greater explanatory value was attributed to the participant analyses. Semantic priming effects on TOT resolution The 2 (Age Group) X 2 (Target Type) X 2 (Prime Condition) ANOVA revealed a marginal main effect of targ et type (by participants), F 1 (1, 65 ) = 3.73 MSE = .048 p 1 = .0 58, r 1 = .23, F 2 < 1 p 2 = 491, where non names (34.4%) were resolved more frequently than proper names (29.2%). There was also a significant interaction between target type and age group (pa rticipants only), F 1 (1, 65 ) = 4.13 MSE = .048 p 1 = .0 46, F 2 < 1 p 2 = 838. Focused contrasts showed that older adults ( p 1 = .008 r 1 =.32 ) had a higher resolution rate for non resolution rates did not diff er as a function of target type ( p 1 = .942).The age group effect was not significant for either target type, p s > .242. Semantic primes did not influence TOT resolution, evidenced by no main effect, F 1 < 1, p 1 = .91, F 2 < p 2 =.391, nor any interactions wit h prime condition, p s > .136. Effects of sylla ble frequency on TOT resolution To see how first syllable frequency influences TOT resolution independent of priming an Age Group X Target Type X Syllable Frequency ANOVA was conducted on the mean TOT resolu tion rate. Only results relevant to the new variable, syllable frequency, are discussed. Syllable frequency had no effect on TOT resolution rate, F 1
111 (1, 64) = 1.19, MSE = .079, p 1 = .280, F 2 < 1 p 2 =.531, and no other interactions involving syllable frequ ency were significant, p s > .289. Discussion Results of Experiment 2 replicated some of the key findings reported in Experiment 1, confirmed a long standing hypothesis regarding the source of TOT states, and offered some new information about the top dow n processing of names and non names from the point of semantic activation through lexical selection. By presenting semantic primes prior to the target question and after TOT states, my goal was to examine how semantic associates enable or interfere with le xical selection of proper names and non names. While semantic primes had no influence on TOTs, for either for non names with HF first syllables. As with Experiment 1, t he specificity of priming effects are useful in identifying structural differences in the representation of names and non names, and how they are uniquely affected by aging. Because person semantics are associated with increased specificity (each person ha s a unique semantic profile), lexical selection of proper names is unaffected by competition from semantically related name targets by increasing activation of alternate lemm as, which suggests that there may be age related changes to either the semantic organization of words or the ability to inhibit semantic competitors. The lack of semantic priming effects on TOT incidence lends additional empirical support in favor of model s that attribute TOTs to a breakdown in lexical to phonological communication. The TOT experience represents a situation where lemma selection has
112 occurred (granting the speaker the strong feeling of knowing the word to be produced), but the retrieval of t he phonological word form is unattainable. As such, it serves as a real world demonstration of the rift between semantic level processing and phonological level processing. Because semantic associates should influence the c features, but not phonological features, the semantic prime was not expected to reduce TOTs or facilitate TOT resolution, as observed in the current data. As such, these results suggest that there is a tangible separation between semantic and phonologica l processing of words, and it is in this rift that TOT states reside. The division between semantic and phonological systems (and its influence on TOTs) appears to be one commonality between the representations of non name and proper names. However, seman tic primes did influence lexical selection but only for older adults name target questions. Further investigation of the semantic priming effect revealed that the reduction in older a incorrect known responses following a semantic prime, as opposed to having more unknown responses. By increasing the likelihood of incorrect alternate answers, the semantic prime (or distract (Ferriera & Griffin, 2003) These findings parallel results from the picture word interference paradigm, where pictures of objects are named more slowly when a semantically related distractor word is visually or auditorily presented with the to be named picture (e.g., Abdel Rahman & Melinger, 2007; Damian, Vigliocco, & Levelt, 2001; Kroll & Stewart; 1994; Sailor et al. 2008; Schriefers, Meyer, & Levelt, 1990;
113 Starreveld & La Heij, 1995, 1996) However, semantic interference effects in the picture word interference paradigm ha ve been observed in both young and older adults, whereas in the current study, only older adults exhibited lexical mis selection following semantic primes and only for non name targets with HF first syllables. Whereas in the picture word interference paradigm responses are slowed due to competition from competitors, semantic interference in our TOT elicitation task represents a non speeded decision to select an alternative lemma, a longer lasting interference effect. The specificity of semantic interference effects observed here suggest s that there needs to be two source s of competition in order for an incorrect lemma to be selected in lieu of the target lemma (e.g., bottom up competition from similar sounding words, and top down competition from semantically related words), and that only older adults are vulnerable to this long lasting interference effect. While proper name retrieval is assumed to involve the same generic st ages as non name word retrieval (semantic activation lexical selection phonological encoding), the current findings challenge the extent to which the semantic organization and lexical retrieval of proper names is really comparable to other word classes Semantic priming (or interference) effects were not found for proper names, for either age group, which suggests that 1) the semantics affiliated with a given person are relatively isolated in semantic space, in contrast to the network of overlapping sem antic features that exists among words, and/or 2) that compared to non names, the selection alternate lemmas.
114 Overall, results from two experiments introduce several w ays in which the representations and processes dedicated to proper names are distinct from other types of words, thereby exacerbating age related declines in retrieval. While the selection of a specific proper name lemma seems to be less affected by interf erence from semantic associates, the selection of a first or last name is vulnerable to competition from similar sounding names once the full name phrase lemma has been activated for production. In the general discussion, I will interpret these findings in conjunction with existing theories special status in the lexicon.
115 Table 4 1. Descriptive characteristics for young and older participants in Experiment 2 Age Group Young Older M SD M SD Age* 19.4 1.6 69.0 4.4 Health Rating (out of 10) 8.3 1.4 8.0 1.5 Education (in years)* 13.4 1.2 17.0 3.1 Vocabulary (out of 25)* 13.1 3.8 19.5 4.3 Forward Digit Span 7.4 1.1 7.2 1.3 Backward Digit Span 5.2 1.3 5.2 1.4 MMSE --28.5 1.1 Note: indicates significant age group differences (p <.001).
116 Table 4 2. Percentage of each response type as a function of age group and target type Age Group Young Older Non Names Correct Known 35.5 51.5 Incorrect Known 16.1 16.0 Cor rect TOT 6.9 10.4 Incorrect TOT 2.1 2.9 Unknown 38.6 17.4 Proper Names Correct Known 25.0 43.4 Incorrect Known 5.5 8.9 Correct TOT 9.0 16.3 Incorrect TOT 1.8 2.5 Unknown 58.3 27.9
117 Table 4 type, prime condition, and first syllable frequency. Age Group Young Old er M SE M SE Non Names Primed HF First Syllable 7.7 1.2 10.3 1.4 LF First Syllable 4.9 1.0 11.4 1.2 Unrelated HF First Syllable 7.2 1.2 7.5 1.5 LF First Syllable 6.6 1.1 11.1 1.3 Proper Names Primed HF First Syllable 9.8 1.5 18.1 1.9 LF First Syllabl e 8.4 1.3 15.1 1.6 Unrelated HF First Syllable 9.5 1.4 17.0 1.7 LF First Syllable 7.8 1.3 15.2 1.6
118 Table 4 function of target type, prime condition, and first syllable frequency. Age Gr oup Young Old er M SE M SE Non Names Semantic Prime HF First Syllable 33.9 2.3 47.2 2.8 LF First Syllable 35.9 2.4 49.6 2.9 Unrelated HF First Syllable 32.1 2.3 53.9 2.8 LF First Syllable 35.7 2.5 52.2 3.1 Proper Names Semantic Prime HF First Syllable 26.3 2.4 42.0 2.9 LF First Syllable 25.7 2.5 44. 2 3.0 Unrelated HF First Syllable 24.8 2.4 40. 4 2.9 LF First Syllable 24.1 2.2 4 7.4 2.7
119 Table 4 type and prime condition. Age Group Young Older M SE M SE Non Names Primed 30.4 4.7 38.0 4.9 Unrelated 31.4 5.7 37.6 5.9 Proper Names Primed 30.6 4.1 27.3 4.3 Unrelated 31.8 4.5 27.0 4.7
120 Give Answ er: K, U, or TOT Give Answer: K, U, or TOT Give Answer: K, U, or TOT Prime before Target Question Prime after Target Question If K: Produce Word Herbivore Next Trial What do you call the type of organism that is anatomically and physiologically adapted to eat plant based foods? Give Answer: K, U, or TOT What do you call the type of organism that is anatomically an d physiologically adapted to eat plant based foods? Unknown Overlooking South Beach, the most expensive condominium in the world is located in which Florida city? Give Answer: K, U, or TOT Overlooking South Beach, the most expensive condominium in the world is located in which Florida city? A vegetarian is a human who refrains from eating meat, whereas a vegan also excludes which other two products from his/her diet? Figure 4 1. Trial st ructure for Experiment 2 A vegetarian is a human who refrains from eating meat, whereas a vegan also excludes which other two products from his/her diet?
121 Figure 4 2. Age differences in TOT incidence as a fu nction of target type and first syllable frequency. 0 2 4 6 8 10 12 14 16 18 20 Non-Name Proper Name TOT Incidence (in %) Target Type TOT Incidence for LF First Syllable Targets Young Older 0 2 4 6 8 10 12 14 16 18 20 Non-Name Proper Name TOT Incidence (in %) Target Type TOT Incidence for HF First Syllable Targets Young Older
122 Figure 4 3. Age differences in correct response rates as a function of prime condition, target type, and syllable frequency. 0 10 20 30 40 50 60 HF Syll NonName LF Syll NonName HF Syll Name LF Syll Name Correct Known Rate (in%) Age Differences in the Unrelated Condition Young Older 0 10 20 30 40 50 60 HF Syll NonName LF Syll NonName HF Syll Name LF Syll Name Correct Known Rate (in%) Age Differences in the Semantic Prime Condition Young Older
123 CHAPTER 5 GENERAL DISCUSSION The goal of this dissertation was to uncover differences between proper names and non names at the levels of semantic activation, lexical selection, and phonological encoding, with the greater goal of understanding of why proper names are particularly susceptible to retrieval failures in older age. Previous theoretical and empirical efforts at understanding the specialness of proper names has largely focused on semantic characteristics. Relative to other words, proper names are inherently more arbitrary and more specific (e.g., Bredart et al., 1995 ; Semenza, 2006; Semenza, 2009), less descriptive and less meaningful (e.g., Cohen & Burke, 1986 Fogler & James, 2007), and are generally encountered less often (Cohen & Burke, 1993; Conley et al., 1999). However, very few studies have direct ly compared proper name retrieval to non name retrieval using a paradigm that enables manipulation of specific features that may d iffer between the word classes. The introduction of semantic and phonological primes at different stages in the retrieval proc ess revealed divergent processing streams associated with names and non names from the moment of lexical selection (when the person has activated the word or name he/she wants to produce) through phonological encoding (when the individual sounds are assemb led for production). Importantly, the retrieval of non names and names was differentially affected by priming and first syllable frequency, suggesting distinct mechanisms to account for age related declines in each word class. In two experiments, the follo wing key differences between proper names and non names emerged: 1) while age differences in TOTs for non names are isolated at the level of phonological encoding, TOTs for proper names may be cause d by retrieval failures at multiple stages of lexical phon ological processing, including first
124 or last name retrieval 2) because proper name TOTs have multiple sources of transmission failures, proper name retrieval is less responsive to phonological priming and is inhibited by phonological neighborhood size, an d 3) due to increased semantic specificity, lexical selection of proper names is a non competitive process. Results from both experiments confirmed a critical role of syllable frequency in the disproportionate age impairment during the retrieval of known names. As expected, there were exaggerated age differences for proper names but only relative to non names with HF first syllables. Words with LF first syllables were as difficult for older adults as proper names, suggesting that age differences in retriev al of non names and names may be caused by different features. For non names, syllable frequency accounted for nearly all of the age related increase in TOTs, as evidenced by the age group X syllable frequency interaction among non names. In Experiment 1, age differences in TOTs for non names were significantly larger for words with LF first syllables compared to words with HF first syllables. In Experiment 2, age differences in TOTs for non names only occurred for words with LF first syllables, consistent with previous research (Farrell & Abrams, 2011). increase in TOTs for non names can be almost exclusively attributed to a breakdown during phonological retrieval, specifically, the retrieval of LF phonological compon ents. been attributed to a long term practice effect: links to sounds that are not used regularly and/or occur in few words grow weaker with disuse, therefore encour aging transmission failures that cause TOTs. In contrast, age differences in TOT incidence for proper names were not dependent on syllable frequency, with older adults universally having
125 more TOTs than young adults for names with HF and L F first syllables. Thus, there must be other factors accounting for age related declines during proper name retrieval. In support of this idea, phonological primes reduced age differences in TOTs for non names but had no effect on age differences for proper name TOTs. One possibility is that there is increased complexity associated with proper name retrieval due to the demands of completing lexical selection (identifying the specific person) and then retrieving the first and last names associated with that lemma. As depicte d in Figure 1 1, there are multiple places wherein a person could fail to activate name label, 2) from the lemma to the last name label, 3) from the first name to its ph onological subcomponents, and 4) from the last name to its phonological subcomponents. Thus, there are at least four locations where activation spreading can fail during proper name retrieval, therefore increasing the likelihood of TOTs The idea that name retrieval is associated with increased complexity, and therefore increased cognitive resource burden, supports previous research indicating more neural resources are involved during proper name recall relative to other words (e.g., Pelamatti et al., 2006; Semenza, 2006; 2009). The cognitive burden of retrieving proper names would be especially taxing on older adults, who have reduced cognitive resources relative to their younger counterparts M ultiple potential loci for proper name TOT s would also explain why first syllable primes were n ot beneficial for reducing TOTs or bolstering TOT resolution of names, for either age group. Because the breakdown that caused the TOT could have occurred at multiple links during the retrieval process, it is less certain t hat the activation from the
126 prime will be able to reach the temporarily disabled link. For example, if the retrieval breakdown occurred in the link between a target lemma and its first name, then activation from a prime sharing only a single syllable would have to traverse the distance between the first syllable node and the first name node and then from the first name to the lemma node. Activation from the prime would diminish over time and distance, so a single syllable may not provide enough information to fully reactivate the link between the name phrase lemma and the first name node. For non names, if the retrieval breakdown occurred in the chasm between the lemma and first syllable, then activation from the prime would only have to spread from the fir st syllable to the target lemma. This might be why priming was helpful for older adults, whose TOTs for non names appear to be exclusively linked to a breakdown during the retrieval of phonological subcomponents. Deactivated links between a proper name le mma and either the first or last name might be the most common source of proper name TOTs, which is why primes containing the full phonology of either the first or last name have been able to reduce TOTs (Burke et al, 2004) and increase TOT resolution for proper names (White et al., in press). It might also explain why people are less likely to have partial phonological information available (e.g., a phoneme or syllable) when having a TOT for a proper name, because they have not yet retrieved the first name have not reach the level of phonological subcomponents. For example, Cohen and Burke (1993) reported that proper name TOTs were associated with complete memory gaps, whereas non name TOTs were more likely to be accompanied by some partial phonological knowledge of the missing word. An interesting question is whether priming
127 of proper names is dependent on the amount of phonology provided or demands an entire name (first or last) to facilitate retrieval. For example, if particip ants are given partial phonology of both the first and last name would that be enough to offset TOTs, or is an entire first or last name needed? names was surprising, given the fact that a n umber of previous studies have shown significant names TOTs using only a single syllable ( e.g., Abrams & Rodriguez, 2005; Abrams et al. 2003; White & Abrams, 2002 ). T resolution for non names was numerically higher in the phonological prime condition compared to the unrelated condition, with an almost 10% difference between the two conditions which is similar to significant priming effect sizes found in previous rese arch In this study, young adults had far fewer TOTs than older adults, so a substantial proportion of youn g adults were excluded from the resolution analyses for failing to have a TOT in each condition. Therefore, the loss of power among analysis for youn g adults may partially account for the non significant priming effect for TOT resolution. However, young adults also failed to show any priming of TOT incidence or correct knowns, which cannot be explained by reduced power, as there were more young adults included in the analysis than older adults. Hence, there must be some fundamental difference in the factors that induce TOTs among young and older adults. This was the first study to examine phonological priming effects on TOT incidence using only a single syllable, as previous studies had used Burke et al., 2004; James & Burke, 2000). One possibility is that younger adults were
128 less familiar with the non name targets, so prov iding partial phonology prior to the question was not enough to forestall TOTs or boost successful selection of the targets However, once a TOT state occurs (and the lexical representation of a word was fully activated), activating a single syllable may p rovide enough information to overcome the TOT, which is why young adults have been able to benefit from single syllable primes in previous research and trended in that direction in the current study. The increased complexity of name retrieval might also ex plain why older adults showed opposing syllable frequency effects for names and non names. The differential influence of syllable frequency on older adults' non name and proper name TOTs can be explained if the effect is assigned to different levels in the retrieval process: facilitation from HF first syllables at the level of phonological encoding of non names, and inhibition from HF first syllables at the level of first name selection of proper names. emma but is unable to retrieve his/her first name, then it is not helpful to have multiple other name options that sound themselves do not denote meaning and are not di rectly linked to the semantic s of a person). The parallel effect for non names is unlikely to occur because words that share phonological properties do not typically share semantic properties, so are not viable candidates for production However, inhibito ry syllable frequency effects were not limited to older adults TOTs for proper names. A relatively unexpected finding was that o lder adults correct names both showed inhibitory sylla ble frequency effects, with fewer correct responses for items with
129 HF first syllables compared to LF first syllables. These findings are intriguing because they imply that phonological characteristics can influence how strongly a word or name is stored in memory. In this case, the possession of a large first syllable cohort reduced the likelihood of completing lexical selection of the target These results are contrary to previous research showing facilitation from high syllable frequency during phonologica lly driven tasks, such as word and pseudoword naming (e.g., Cholin et al., 2011; Macizo & Van Petten, 2007). The current data suggest that having many first syllable neighbors can be counterproductive during all stages of lexical access prior to p honologic al encoding, at least when there is uncertainty of the word to be produced. the non name targets was significantly weaker than that of older adults. If know ledge of the non names was characterized by familiarity with the target concept but uncertainty with the word itself (i.e., before lemma selection) then it might create a situation wherein having a numbe r of similar sounding words hinders access to a spec ific word For older adults, interference from similar sounding proper names may have occurred during the retrieval of the first name, after successful selection of the lemma associated with the person. Overall, it appears that facilitation from HF syllab les only occurs when it is at the level of phonological encoding, when the person is cer tain of the word to be produced, such as names and previous studies on word or pseudoword naming tasks which do not demand lexical selection In other cases, when a person is less certain of a word (i.e., young adults retrieval of non names), or when
130 proper names), it becomes more problematic to have similar sounding words and names. Although these assumptions are speculative, it might be a useful future research endeavor to see if the beneficial effects of phonological frequency on word production reported in previous research can be reversed when participants are asked to retrieve words from memory when uncertain of correctness of their response. For example, would high syllable frequency interfere with naming latencies when participants are asked to name pictures of uncommon objects whose na mes are known but not produced often? Or, would syllable frequency negatively influence recall of newly learned words that have just entered the lexicon because they sound like other existing (better known) words? It is important to note that the dispropo rtionate age impairment for names and its interaction with syllable frequency is 1) specific to TOT incidence, and 2) cannot be explained by age differences in familiarity with the targets or the number of opportunities to have TOTs. First, ifficulty with proper names was only expressed in the TOT incidence analyses; older adults knew more correct responses and had fewer unknown responses for proper name questions relative to young adults Thus, the particular age impairment for proper names is isolated to the level of phonological retrieval, i.e. influencing only TOTs, when older adults are attempting to retrieve names the proper name targets granted th em fewer opportunities to have TOTs; one can only have a retrieval failure if there is an existing representation that is trying to be accessed. However, identical patterns emerged when unknown answers were excluded ( including all trials for which a retrie val attempt was made) and when TOTs were calculated as a
131 proportion of known answers ( including all trials for which the correct answer was known ) suggesting that age difference discrepancies as a function of target type cannot be explained by familiarity Further, the interaction between age group, target type, and syllable frequency remained unchanged when the analysis was conducted using only targets matched on first syllable frequency. Thus, even when looking at the identical set of syllables, high syl names but increases their TOTs for proper names. Beyond the organization of lexical phonological links, Experiment 2 revealed intriguing differences between names and non names in terms of the flow of act ivation between the semantic and lexical layers. A critical question was whether lexical selection of proper names is a competitive process, as is assumed by most models of non name word production. Based on previous research, there were a priori reasons t o suspect that proper name selection would be more vulnerable to competition from semantic competitors due to a close knit distribution of proper names within semantic space (e.g., Burgess & Conley, 1998; Conley et al., 1999).There were also reasons to sus pect that proper names would be less vulnerable to lexical competition due to the individualized nature of a person's semantics ( e.g., Semenza, 2006; 2009) Because names but not for name s, results from the current research confirm the latter perspective and suggest that the semantic representations of names are characterized by less semantic overlap among semantic associates, resulting in a non competitive lexical selection process. Olde selection in that they showed an increase in incorrect responses following a semantic prime but
132 not a change in unknown responses. The incorrect selection effect can be explained by dif ferences in the relative activation levels of viable lemmas in the primed versus unrelated condition. In the unrelated condition, each question is read as its own entity, triggering activation of the semantic features described in the question and allowing the participant to select the specific lemma that best embodies the set of activated semantic features. However, when the prime is presented in the pre target question, participants are forced to engage in deep semantic processing of the prime word and it s meaning, representations of other semantically related lemmas. The subsequent target question then activates a set of semantic features that overlap closely with the prime but are not identical. Because of the proximity between the prime and target questions, there is not enough time for the activation of competing lemmas to completely subside. Hence, the activation of shared semantic features (provided in the targ et question), coupled with (provided by the prime question), allows for an incorrect word to be chosen for production. Participants are only susceptible to this interference effe ct prior to the point of lexical selection of the target (TOT incidence) which is why semantic primes do not influence TOT resolution, where they have already chosen a particular lexical entry for production. However, the semantic interference effect was not ubiquitous O nly older adults experienced the lexical mis selection effect and only for non names with HF first syllables. Although young adults have shown semantic interference in other research methodologies, t he type of semantic interference reporte d here differs from other
133 paradigms in that it is relatively long lasting ; activation of alternate lemmas must be maintained at least several seconds in order to prevent selection of the target. Semantic interference effects in the picture word interferenc e paradigm for example, are much more ephemeral, so much so that they disappear when the timing of the distractor/picture presentation is changed by mere milliseconds Further, semantic distractors in those studies generally slow production latencies but rarely cause the participants to produce the wrong picture name. Hence, there is something methodologically unique about the type of distraction provided by the semantic primes in this study that allow s for the wrong name to be chosen. Based on the selecti vity of when interference occurs, it seems that competition must come from two sources in order for lexical mis selection to occur: semantic interference from activated competitors, and phonological ambiguity from HF first syllables. It was not activation of the prime itself that caused older adults but not young adults to produce more incorrect responses following prime questions as both age groups were equally likely to give non name primes as the answer to non name target questions. Therefore, semantic primes must be encouraging older adults to produce alternate words that had been activated in the prime question. Words with HF phonology might also have more semantically related words relative to words with LF phonology, rendering them more vulnerable to the mis selection effect lasting semantic interference effect can be explained by two different sources of age related change in the cognitive and language profiles of older adult s declines in inhibitory con trol ( e.g., Hasher & Zac k s, 1988; Zacks & Hasher, 1994 ), and gains in semantic word knowledge (see Burke & Shafto, 2008 for
134 review ). The ID hypothesis is a resource reduction theory of cognitive aging that explains age related deficits across multiple cog nitive domains to a common agent: reduced ability to ignore or suppress irrelevant stimuli from entering attentional awareness. The ID hypothesis has been used to account for a number of documented age related changes to language processing, including incr eased word length and tangential speech during discourse production, and increased interference from distracting material during comprehension (see Abrams & Farrell, 201 1 ; Burke & Shafto, propensity toward semantic interference observed in the current study could be caused by an inability to inhibit or suppress previously activated lemmas while reading the target question. Because of this sustained activation of the alternate (incorrect) le mma s they are unable to fully process the target question and fail to detect the introduction of a new concept that they would have otherwise known (in the un related condition ). Alternatively, older adults may be more inclined to semantic interference bec ause they have superior semantic networks relative to young as measured by vocabulary scores and general knowledge tests ( e.g., Verhaeghen, 2003; Ackerman & Rolfhus, 1999) In order to see semantic priming or interference, the semantic networks dedicated to the target and prime must include shared semantic representations. If older adults have better knowledge about the meanings of both the prime word and the target, they would be more likely to activate the semantic features shared between the words, allo wing for lexical selection of the prime and/or related lemma to supersede selection of the target. For this same reason, older adults might activate more competing lemmas while reading the prime question relative to young adults, making target selection mo re difficult.
135 Neither age group experienced semantic interference effects during the retrieval of proper names, suggesting that proper names are either less likely to share semantic features (therefore preventing the spread of activation between proper na mes at the semantic level) or that the occurs independent of the relative activation levels of other names The former point is somewhat counterintuitive because obviously people do possess overlapping semantic features; they share common occupations, hobbies, acquaintances, events, and projects. And, both types of target questions were designed to engage those features that were shared between target and prime. So, why then would activation from the semantic prime influen ce lexical selection for non names but not for names? One obvious concept is that the representation of a person demands increased specificity, both at the level of semantics (each person that we know has a completely unique constellation of semantic featu res) and at the level of name retrieval (there is always only one name lemma that matches those semantic features). There is less competition during the selection of name because there is no other word or label to communicate the same level of specificity. In contrast, words often have synonyms or very closely related alternate word options that can be produced in place of a given target word, making In sum, data from two experiments unveil differences in the r epresentation and processing of proper names at multiple levels in the language network. Some of these differences seem to exacerbate the negative effects of age on the retrieval of proper names compared to other words. The lexical phonological representat ion of names is characterized by added complexity due to the necessity of retrieving first and last
136 names, many of which are shared by other people that are known. Because of this additional level of complexity, there are multiple places where transmission failures can occur, which increases the likelihood of TOTs, especially for older adults The specificity demanded by proper names (both in the semantic profiles associated with a given person, and the specificity of the name label used to describe that pe rson) causes lexical selection of proper names to be less dependent on competition. While the focus of this research was to understand how names and non names are differentially portrayed in the lexicon, these data also revealed some insight abut general l anguage processing, including the opposing influence of syllable frequency during lexical selection versus phonological encoding, and the extent of interactivity between semantic, lexical, and phonological characteristics during word retrieval and producti on. Conclusions from this research are limited by the fact that the proper name and non name stimuli were not perfectly matched on familiarity, so we cannot completely rule out the possibility that lack of familiarity with the proper name targets contribut ed to the increase in retrieval failures for names (independent of age). Additionally, the measure of syllable frequency used here represents the combined frequency of words that contain that syllable in the onset position, as reported by the CELEX databas e While CELEX does contain a few proper names within the database, many of the names used in this study were not found in the database, so their syllable frequencies came from a corresponding non name with the same first syllable. At any rate, it is not k nown whether a HF syllable in language is necessarily a HF syllable among names. In future research, I hope to disentangle this issue to see whether inhibitory effects of high syllable frequency among proper names stem from a name sounding like many other
137 words (non names) in the lexicon or from names sounding like many other names. U spreading activation among similar sounding names influences retrieval. The finding that older unexpected and theoretically valuable contribution to models of speech production, which have generally assumed distinct semantic and phonological systems. One unexplored possibil ity is that HF first syllables are more likely to be morpheme units that denote some level of semantic meaning. As such, there may be more semantic commonalities among words with HF first syllables than words with LF first syllables, allowing two sources o f interference (semantic, phonological) in the former case. Because I did not control for morpheme usage in the current stimuli, it might be a useful future research endeavor to see whether the relationship between syllable frequency and semantic interfere nce is driven by the morpheme status of the syllable Finally, it must be pointed out that the conclusions I have drawn about age related changes to proper name retrieval refer only to their retrieval of famous names. While I have generally used the term of course many other types of proper names (place names, event names) that may enable different storage and processing mechanisms than that of people's names. At this point it is largely unknown whet her these other types of proper names are equally vulnerable to retrieval failures as people names and whether they are differentially represented at the semantic and phonological levels. It would be particularly useful to contrast proper names that have t o share lexical nodes with other names (i.e., other Julies, other Smiths) to those that do not involve this particular constraint (Afghanistan,
138 Paris). Comparing age differences for words, completely unique proper names, and shared proper names would revea l the importance of name sharing (and phonological during name retrieval To the same end, diary studies have suggested that acquaintance names (not famous names) are the most common source of TOTs amon g older adults, although the retrieval of acquaintance names has yet to be examined within the context of a laboratory study. There are reasons to suspect that acquaintance names would behave differently than proper names due in part to the specific featur es that were uncovered in the current study. First, acquaintance names may be more common than famous names, so there may be increased sharing of first/last names and hence increased interference during retrieval. Further, many of the famous names used in TOT studies represent the most iconic version of a specific first or last name, so they may be less vulnerable to phonological interference than would acquaintance names. Globally, I plan to expand on this line of research by investigating how the fragilit representation is linked to breakdowns during the encoding and/or learning of new names. Are unique names better learned because they do not compete with existing proper names that are already associated with other people? Are distinct n ames more memorable because there is consistency in the mapping of phonology and semantics, whereas non names refer to many different semantic instances ? For example, when most people hear the name Barack they think of one semantic instance of that name ( t he first African American president of the United States ) whereas Betty might refer to any number of famous women with that name. Further, how do metacognitive factors contribute to difficulty with learning names? Answering these question may lead to a
139 bet ter understanding of how names are differentially represented in semantic on.
140 APPENDIX A NON NAME TARGET, PRIME A ND UNRELATED QUESTIO NS Non Name Target Target Question Phonological Prime Question Semantic Prime Question Unrelated Prime Question Acronym What do you call a word or abbreviation, such as CEO, that is formed by taking the first letter or letters from each word in a phrase? Which two variables are used to calculate bo dy mass index to get a relatively accurate measure of one's body fat? In public transportation, what do the initials ETA stand for? Increased incidence of forest fires is a surprising result of __________ _________, the rise in the temperature of Earth d ue to Greenhouse gases? Alchemy What medieval 'science' was concerned with the conversion of the base metals into gold and the creation of elixirs for youth and immortality? What does it mean to dine alfresco ? Despite its negative connotation, practition ers of witchcraft describe themselves as healers, similar to which type of spiritual leader thought to treat illness by contacting the spirit world in trance like rituals? Seaweed is often used in wrapping which kind of Japanese food consisting of cooked rice, raw fish, and various other edible ingredients ? Allegory What is the term for an artistic device used to express an idea through symbolic images or words, such as the grim reaper to represent death? Mother Theresa is thought to be one of the most a ltruistic humans in the 20th century, and in 1979 was awarded which Swedish prize for peace? Which classic fable tells the story of a race between a slow, steady creature and a fast, lazy creature, warning the listener about the dangers of overconfidence? What is the term for a drug, such as marijuana or tobacco, that is thought to create the behavioral doorway into the use of more serious drugs?
141 Anagram What type of word play involves rearranging the letters in a word to create a new word or phra se, using all letters exactly once? In the NFL, how many yards' penalty would a player receive if the referee sees him antagonize or taunt the opponent? What is the name of the word puzzle where the goal is to fill in white squares with letters that form w ords by solving clues which lead to the solutions? The 'Speaker' is the title given to the presiding leader of which legislative body? Anarchy What is the term for a society without any form of organized government, also used to describe political disord er or lawlessness? The phrase anecdotal ________ refers to testimony or data based on the personal stories or reports of a few individuals? Nihilism, the philosophy that negates the idea that life has meaning and challenges the existence of morality, is of ten associated with which German thinker? What do you call the style of jeans or trousers that become wider from the knees downward and were fashionable in the 1960s and 70s? Annulment What is the term for a legal or judicial procedure that declares a ma rriage invalid or void from its beginning, as if the marriage had never occurred? Which tune played during the famous Disney World musical boat ride has been voted the most annoying song in the world? In 2010, which country had the highest divorce rate in the world and is also geographically the largest country in the world? What do you call the person who deals directly with customers at the bank, and whose responsibilities include check cashing, deposits, and cash withdrawals? Anthrax What is the word f or the infectious disease that is transmitted by handling contaminated materials, that caused a crisis in the United States in 2001 when it was sent to multiple public figures in envelopes? What is the most anterior lobe in the human brain? In 2003, U.S. intelligence uncovered a terrorist plot to release cyanide gas into the New York City subway system, organized by which militant Islamist group? What plant eating insect is the larval form of a butterfly in its undeveloped early stages?
142 Arson What is the crime of intentionally or maliciously setting fire to structures, wild land, or other property? What is the name of the film review wesbite that uses the ripeness of tomatoes for its arbitrary rating scale? Less than ____% of people have pyromania, the imp ulse control disorder characterized by the need to start fires to relieve tension or for gratification. Which 2006 'mockumentary' is about a spoof reporter from Kazakhstan and his ridiculous capers in the United States? Audiologist What type of health pr actitioner specializes in evaluating and treating individuals with hearing loss or dysfunction in the vestibular system? What is the Earth's highest mountain, luring many audacious climbers and mountaineers to the Himalayas each year? Optometry is a healt h care profession whose practitioners are trained to treat symptoms and diseases of which sensory organ? In the 1960s, protest demonstrations became highly popular in the U.S. for a variety of social groups trying to voice their disapproval of which war? Barnacle What do you call a small, hard shelled marine animal that attaches itself in clusters to submerged surfaces in shallow tidal waters, and resembles a small crater once attached? In a barter system, goods or services are directly exchanged for other goods or services, without the use of what intermediary medium? What is the term for a substance or food that increases sexual desire, such as the edible shell covered mollusk called an oyster? Which youth organization for girls places an emphasis on prac tical outdoor activities, including camping, woodcraft, aquatics, hiking, backpacking, and sports, and is known for its annual cookie fundraiser?
143 Beaker What type of laboratory glassware is used to stir, mix, and heat liquids, and is generally shape d like a cylinder with a flat bottom and small lip to aid pouring? One of the most famous upsets in sports history occurred in the 1980 Winter Olympics when the Soviet hockey team was beaten by which underdog nation? What do you call the narrow tubular sec tion at the top of a laboratory flask, also a word for a human body part? What type of alternative medicine involves the manipulation of very thin needles inserted into the skin, a painful sounding treatment that most actually find quite pleasant? Boomer ang What is a flat, curved, wooden flying tool used as a weapon or for sport, that is designed to return to the thrower once it's in the air? The tradition of booing a performance has its roots in which ancient culture's annual festival where playwrights w ould compete to determine the best tragedy? The Frisbee was trademarked by which U.S. toy company that also owns the rights to the Hula Hoop, Slip N Slide, and Hacky Sack? The panda is an endangered bear species, with fewer than 3,000 animals living in wh ich Asian country? Cadaver What do you call a dead human body that is used for dissection and medical research? What is the name of the social networking site where users connect with one another through 'tweets'? What is the largest bone in the human ske leton, located in the upper leg? Around which body part are you supposed to tie a string in order to remind yourself of a task you intend on doing in the future? Calligraphy What is the art of elegant, decorative lettering, that is often used for wedding and event invitations, inscriptions, and formal documents? Traditionally, volunteers in the Peace Corps have to commit to how many years of service abroad? According to popular belief, which type of professional is notorious for having sloppy, unclear ha ndwriting? Certain plants rely on the scatter of their seeds though wind or animal transport to accomplish which process?
144 Capillary What is the smallest of the body's blood vessels that receives blood from the arteries and arterioles and transports blo od to venules? The Saffir Simpson scale is a 5 level categorical system used to rank the strength of what weather event, whose peak Atlantic season occurs in August and September? In the circulatory system, do veins transport blood towards or away from th e heart? Which species of freshwater fish resides in South American rivers and has the reputation for an insatiable appetite for meat? Caravan What is the word for a group of travelers on a journey, typically through the desert or on a trade mission, also used to describe a covered vehicle used as traveling living quarters? What is the legal term for the killing of a human being without intent, such as when a patient dies due to the careless negligence of a doctor? What type of nomad, like the first moder n human, relies exclusively on wild plants and animals for food as opposed to domesticated species? The most famous use of the diamond gemstone is to adorn which piece of jewelry that symbolizes a future union? Carcinogen What is the name for a substance or agent that directly produces cancer, such as tobacco? Carving a pumpkin creates which type of Halloween ornament? The inhalation of asbestos can lead to which type of cancer? How does one describe a very bright hue or shade, such as those used in a high lighter, that is also used in high visibility clothing so that the wearer is easily distinguishable from the background?
145 Christening What is the religious ceremony (another word for baptism), that involves 'naming' an infant within the religion and a w ater ritual that symbolizes an infant's admission into the faith? Which form of precipitation occurs when water begins to crystallize within the Earth's atmosphere and most often occurs during winter months? In the Jewish faith, what is the female equivale nt of a Bar Mitzvah the ceremony that symbolizes when a person becomes responsible for his own actions from a religious perspective? Traditionally, an overcoat is made of a heavy, warm fabric intended to serve as the outermost garment, and is designed to extend below which body part? Contraband What is the term for items that are illegal to be possessed or sold, typically associated with goods that are smuggled or traded illicitly? What famous North American landmark is constantly moving backwards becaus e of the millions of gallons of water that move over it every minute? The term bootlegger started when horsemen would use their tall boots to hide pints of alcohol and became popular during which reform period in United States history when the manufacture and sale of alcohol was made illegal? Which glorious event occurred in Berlin in 1989, symbolizing the reunification of East and West Germany? Contraception What one word term refers to the prevention of pregnancy through various natural and intentional techniques, such as breast feeding, hormone pills, or barrier methods? What is the term for the concise written records of a meeting that summarize the events of the meeting and any decisions that were made? A vasectomy is a surgical procedure that preven ts the release of ______ when a man ejaculates. In which famous play do both lead characters, star crossed lovers from opposing social groups in Verona, Italy, tragically die after a series of hasty decisions?
1 46 Coroner What is the name of the public offi cer whose primary functions are to investigate the origin of human deaths, issue death certificates, and maintain death records? What is the method of transmitting information through a series of on off tones, lights, or clicks that allows people to effici ently correspond over radio? What procedure involves slicing open and separating body parts to investigate a body's internal structure, often used for scientific investigation or autopsy? What is the term for a survey of public opinion from a particular sa mple that measures things like the percentage of people that disapprove of a president or certain policies? Crucifix What do you call the image of Jesus's body hung on a T shaped wooden structure, often used in religious jewelry and artwork, that is dist inct from the image of the structure with no body? What is the largest organ in the human body, crucial for protecting the body from most physical and environmental factors? The resurrection of Jesus is honored on which holiday, marking the end of Lent? D espite media reports, job satisfaction is not strongly correlated to which outcome, a measure that reflects the efficiency of production output? Demographics What is the word for the statistical characteristics of a population, that typically include info rmation about a region's gender, age, ethnic, or financial profile, often used in sociology and marketing? What is the name of the delicate fabric, not silk, that has a glossy surface and is often used in women's lingerie and evening gowns? The Unites Sta tes Constitution mandates that a census be taken every ________ years to track changes in the size of the population as well as its social and ethnic makeup. The phrase 'slippery ____' is used to describe the idea that a relatively minor first step can lea d to a chain of events that result in something significant.
147 Diameter In geometry, what word represents a straight line through the center of a circle? What is the name of the most common form of traumatic brain injury, whose symptoms include temporary memory loss, nausea, headache, and dilated pupils? Circumference refers to the length of the perimeter of which geometric shape? Which basic mechanical device is shaped like a helix and made of elastic material, and is used to absorb shock, such as in mat tresses and upholstery? Diaphragm What sheet of muscle extends along the bottom of the rib cage separating the chest and abdominal cavities, that singers use to control their breathing in a performance, and also causes hiccups when it spasms? What is the name of the medical removal of tissue from a living subject to diagnose the presence of an infection or pathology? What do you call inflammation of the larynx, which causes hoarseness or a loss of voice and is often associated with singers? Aside from the treadmill, what is one of the most popular aerobic machines used in gyms, designed to simulate stair climbing, walking, or running without causing excessive pressure to the joints? Dowry What is the term for the money, goods, or property that a bride bri ngs to her husband at marriage? Considered dowdy by previous generations, which length/style of skirt has become a hot fashion trend in recent years? What is the legal contract that names who will manage an estate and receive inheritance of property at dea th? A playful piano tune by Chopin is nicknamed the ______ Waltz, although it typically lasts between one and a half to two and a half minutes.
148 Ecosystem What is the term for a contained biological community (not habitat) consisting of living organisms and their nonliving surroundings, such as a coral reef or tundra? What is the term for the type of motor vehicle that uses two or more power sources, such as a Prius, that has grown in popularity due to the need to economize on fuel prices? The planet's biosphere is thought to have evolved with a process that began approximately 3.5 ________ years ago. What do you call wood from a variety of species that is suitable to be used by a carpenter, often supplied to hardware stores or directly to construction sites? Embryo What do you call an animal in the earliest stages of its development in the uterus, specifically in humans from conception to about the eighth week? Which developmental disorder typically diagnosed in childhood is characterized by impaired s ocial interaction, an inability to empathize with others, and/or restricted or repetitive behaviors? Skipping meals during pregnancy may harm the development of a human fetus and increases the risk of which condition, when a baby is born before 37 weeks o f gestation? Due to serious budget shortfalls, in 2010, President Obama halted NASA's plans to revisit what celestial body, the Earth's only satellite? Equestrian What is the term for a person who skillfully rides on horseback, typically for recreation, c ultural display, or for competition in horse shows? Although one should not equate the test with intelligence, which test taken at the end of high school has been shown to predict academic achievement in college? What is the average weight for a jockey, t he small athlete that rides horses during horse races? Which measure used in psychology research is used to assess how quickly humans react and respond to a stimulus? Euphemism What is a word or phrase that is substituted as a less distasteful term for on e considered offensive, such as 'passed away' for 'died'? What is the second most popular illegal drug in the U.S., an inhaled stimulant that produces a feeling of alertness, euphoric mood, and increased sexuality? What is the French inspired term for a fi gure of speech that is designed to be interpreted in multiple ways, such as a sexual innuendo? What term describes the celebration of a 100 year anniversary, usually held for a place or organization?
149 Genealogy What is the study of family histories that in volves tracing a family's descent from ancestors? Which 'Opposites Attract singer was also the most genial of the original American Idol judges? What is the name of the NBC documentary series where a celebrity goes on a journey each week to trace his or h er ancestral lineage? Most pencil erasers are made from a synthetic version of which material? Genocide What do you call the deliberate and systematic destruction of a particular ethnic, political, religious, or cultural group? What type of shark is the largest living fish species, a 'gentle giant' that mainly feeds on plankton? What was the term used to describe the massacre of approximately six million European Jews during World War II, a program of destruction ordered by the Nazi leader Adolf Hitler? What adjective is used to describe a creepy foreboding of an upcoming harmful event, such as dark clouds hovering in the distance? Haiku What is the name of the unrhymed Japanese poem having 3 lines containing a set number of syllables, usually on some su bject in nature? Which hygienic practice is cited as the second best way to prevent flu, after the vaccine? What European island nation is credited with developing the limerick, a witty and humorous poem with a strict rhyme structure? The office furniture used to organize and store paper documents in labeled folders is called a ________ cabinet. Harem What word refers to a group of women that belong to one man, typically associated with Eastern and Middle Eastern cultures? What do you call the top ranking employee of the White House Office, a harried, stressful position as the President's closest advisor? In which holy Islamic text is a man's possession of a concubine permitted under strict circumstances? In a newly popular winter celebration, party goers are required to wear which type of 'ugly' garment?
150 Herbivore What do you call the type of organism that is anatomically and physiologically adapted to eat plant based foods? Hurried woman syndrome is a new term used to describe the weight gain, low sex drive, moodiness, and fatigue symptoms that accompany a busy stressful lifestyle, and over time can lead to long term __________. A vegetarian is a human who refrains from eating meat, whereas a vegan also excludes which other two products from his/her di et? Overlooking South Beach, the most expensive condominium in the world is located in which Florida city? Hermaphrodite What do you call an organism that has both male and female reproductive organs? What term is used to describe a condition where a body part continues hurting for a long time, maybe even years? Known for her boyish figure and short pixie cut, which English supermodel made androgyny fashionable in the 1960s? According to the common superstition, breaking a mirror will cause how many years of bad luck? Hieroglyphics What is the name of a well known ancient Egyptian picture writing system? What cartoon starred a hyperactive, freckle faced boy with a cowlick and a penchant for mischief? How many vowels are there in the modern English alphabe t? A delirious mental state and tingling of the limbs is a sign of which medical condition, caused by loss of body fluid? Hypochondria What term describes the condition of being unnecessarily anxious about one's health, or the state of suffering imaginar y illnesses? What popular sports drink was developed by University of Florida researchers to hydrate athletes? What do you call the type of neurosis or anxiety disorder defined by a persistent and often irrational fear of an object or situation that is gen erally disproportionate to the actual danger posed? A stiletto is which type of weapon with a long, slender blade?
151 Incubator What is the term for a device designed to maintain optimal temperature, humidity, and air conditions used within science to grow cell cultures, or for medicine to care for premature babies? Named after the winged bug, what 'effect' refers to the phenomenon whereby a series of seemingly inconsequential events snowball to create a large result? A thermostat is used to maintain a syst em's _________ at a specific set point. What is the lowest satisfactory letter grade according to most American schools? Iota Originally a Greek letter written in subscript, what word is now used to mean a very small amount? Common table salt is formed by an ionic bond between which two chemical elements? In cooking, what is the term for the very small amount of a substance that can be taken between the thumb and forefinger, or the equivalent of two smidgens according to some modern cooking companies? Wha t is another name for table tennis, a sport where players hit a lightweight, hollow ball back and forth across a net ? Javelin What is the name of the slender metal shaft that was once thrown as a weapon of war and is now thrown for distance in a field ev ent? Which U.S. city is known for its jazzy music scene, unique cuisine, and its annual Mardi Gras festival? In Florida, what type of reptilian game is hunted with a spear during nighttime hours when firearms are prohibited? In bowling, a perfect game is a chieved when the bowler reaches the highest possible score of 300 points by rolling a ________ on every frame. Juniper What is the name of the evergreen plant with needle like leaves, whose blue colored berries are used to flavor foods, such as the popul ar liquor that is often served with tonic? Although technically different strands of the same fruit, the peach is thought of as the juicier, fuzzier version of which fruit? Aside from its popularity in Christmas songs, holly is used to make which type of c affeine rich herbal tea from South America that is served with a metal straw? A regional airport is one that lacks customs or immigration facilities, so it cannot offer which kind of flight?
152 Kaleidoscope What is the name of the tubular viewing toy that pr oduces symmetrical designs through an arrangement of mirrors and chips of various hues? One of the most collegial ways to decide on a topic or select a leader is through ________ ballot, where each vote is anonymous. Which mnemonic aid is used to remember the sequence of hues in a rainbow? Which lightweight but strong metallic element is often used to make the head of a hammer? Lentil What is the name of the edible plant that comes from the bean family and originated on the Indian subcontinent, whose seed s can be cooked and eaten and form the base of a popular vegetarian soup? __________ lending refers to the practice of granting loans to individuals who may have difficulty making the repayment schedule, a major factor in our recent economic recession. Whi ch type of Middle Eastern spread is made from ground chickpeas blended with tahini, olive oil, salt, and pepper? What do you call the form of communication designed to persuade the attitude of a community toward some cause or position, often seen as a form of political warfare? Luminaria What do you call a small torch placed inside a paper bag that is usually arranged with others to form a display during holidays? What is the name brand of the popular spray that is used to lubricate hinges and locks and u nbind stuck bolts? Although most are made of paraffin, a candle can be made with which natural product, procured from the hive of a flying insect? The distance of lightning can be calculated by the interval between when the lightning is seen and when the t hunder is __________. Malevolence What noun describes the quality of having or showing vicious ill will or hostility towards someone, or the desire to do evil to others? The opposite of malignant, what do you call the type of tumor that lacks the abilit y to spread? What is the term for an unlawful act that is motivated by the perpetrator's hatred for a specific social group? Which Ivy League university near Boston is considered one of the world's most exclusive, with the second lowest acceptance rate in the country?
153 Marinade What do you call a savory sauce in which meat, fish, or a vegetable is soaked before cooking to enhance the flavor? You are legally required to report a change in marital status to which government agency responsible for administeri ng retirement and disability benefits? Which type of salad dressing is a mixture of olive oil and vinegar, usually flavored with herbs and spices or other ingredients? According to a British research team, which sound do people find most repulsive, a noise made by humans and is typically preceded by nausea? Marsupial What is the name of the order of warm blooded animals including koalas and opossums which carry their young in an abdominal pouch? Which 2010 film about an English folk legend known for 'rob bing from the rich' saw only marginal success at the box office? What do you call a baby kangaroo? Who is credited for inventing the first working telephone, receiving the first patent in 1876? Martyr What do you call a person who suffers persecution, off ers a great sacrifice, or voluntarily dies for refusing to renounce, or accept, a belief or cause? Which 2006 film about the Irish American organized crime scene won Best Picture and featured a marvelous ensemble cast including Leonardo DiCaprio as the und ercover cop Billy Costigan? Who is the patron saint of Ireland, whose legacy is celebrated around the world every year with green beer and parades? What is the name of the impressive wave of protests throughout the Arab world in 2010, noted for its use of social media to spread the word? Mercenary What do you call a professional combatant that takes part in armed conflict strictly for financial gain as opposed to having a direct interest in the war itself? What do you call the murky water that results from the mixing of sea water with fresh water? Which type of soldier in the U.S. Navy is a principal part of the special operations force and has a reputation for enduring one of the toughest training regimens in the world? What do you call a wistful, visiona ry fanta sy, especially one of hopes or ambitions imagined as coming to pass, that occurs while awake?
154 Meteorology What do you call the scientific study of weather and the atmosphere, whose practitioners are often employed by television news programs? What Indian salutation has been given a more meaningful association in western yoga, where it is exchanged between the teacher and students? What is the scientific instrument used to gauge atmospheric pressure and forecast the weather? Which original action f igure of a 'real American hero' is now a collector's item and is one of the most expensive toys in the world? Migration What is the term for the annual seasonal journey made by many bird species in response to changes in food availability, habitat, or wea ther? What is another nickname given to the Minor League Baseball league that started with a joke that players were being grown like corn? What is the smallest known bird species who has the ability to hover in the air when in flight by rapidly flapping it s wings? Which term is used to describe an abandoned village, town or city that was once populated but is now eerily empty, as if occupied by spirits? Misdemeanor What is the name of a 'lesser' criminal offense, such as petty theft or trespassing, with a relatively easy punishment, but is still more serious than an infraction such as speeding? In current popular culture, what do you call someone who looks so similar to a person that they could be mistaken as twins? What is the most prevalent type of felo ny conviction in the United States? What is the name of the Scottish apparel worn by warriors that resembles a knee length plaid skirt? Mitochondria In cell biology, what is the 'powerhouse of the cell', the structure that produces energy? What do you cal l the mighty group of Norse warriors who raided wide areas of Europe and Asia and are often depicted as wearing helmets with two horns? The nucleus of the cell is home to which helical structure that contains the cell's genetic material? How many keys are on the average piano, 88 or 22?
155 Molasses What is the name of the thick, brown substance that resembles honey and is produced during the refining of sugar or by boiling down sweet fruit juice? What type of malicious computer program can replicate itself and spread from one computer to another? Which state in the Northeast is the largest producer of syrup in the U.S., making 5.5% of the global supply of the sugary substance used to adorn pancakes? Which Seattle native that died at the age of 27 is widely regarded as the greatest electric guitar talent of all time? Monastery What is the name of a building or group of buildings that houses individuals who have taken a religious vow and includes a room dedicated to prayer? Which dinosaur with a monstrous hea d and long tail was one of the largest known land predators and has become the best recognized dinosaur species in popular culture? Although the term can be applied to the residence of any religious order, a convent is generally taken to describe the livi ng quarters of which type of spiritual person? Which measure of typing speed is used as an indicator of how adept a professional secretary is at his/her trade? Monogamy What term refers to the practice of having one sexual partner or spouse at a time? Wha t is the term used to describe political ads that are designed to malign the other candidate by exaggerating or bending the truth? What do you call a gentleman above the age of legal adulthood who has never taken nuptial vows, and if past a certain age is often attributed to having a fear of commitment? The violin comes from which class of musical instruments, which also includes the guitar, bass, cello, and harp?
156 Monogram What is a symbol or logo that is made by combining the initials of an individual o r company and is often used to adorn stationary, luggage, or clothing? What does GDP stand for, the monetary value of all the finished goods and services produced within a country's borders in a specific time period? Which French fashion label is known fo r the embroidery of its initials, LV, sewn into its luxury trunks, bags, totes, and apparel? What is the name of the incredible waterfall located in southern Africa on the Zambezi river? Mutiny What is the term for a conspiratorial revolt (not a coup) aga inst authority, such as the crew overthrowing the captain of a ship? On which continent is it a common practice to mutilate a female's genitals for non medical reasons? What Caribbean nation was formerly a French colony before starting the most successful slave rebellion known to history? Which Oliver Twist author, considered the greatest novelist of the Victorian era, was known for including passages in his writings that were designed to subtly satirize British aristocratic snobbery? Nostalgia What word is used to describe the sentimental yearning for a past experience or time period, often in an idealized form? A recent survey found that ________, individuals who reject the idea of a God, are actually among the most knowledgeable about world religions. W hat style of book is dedicated to an author's reminiscence about his or her own life and experiences? What do you call the harmonious ensemble of string, brass, woodwind, and percussion sections, sometimes consisting of over 80 musicians? Obituary What s ection of the newspaper reports the recent death of a person, typically along with an account of the person's life and information about the upcoming funeral? What is the well known list of biblical principles that believers in Judaism and many forms of Ch ristianity are expected to obey? A eulogy is a tribute speech that praises a person's life and accomplishment s and is usually a part of what type of service? What sleep disorder characterized by abnormal pauses in breathing during sleep often goes undiagno sed because people ignore the symptoms?
157 Parasite What is the term for an organism that lives on or in another, obtaining nourishment from it? What is the common name of the substance that is injected into the skin to paralyze facial muscles, thereby reduc ing the appearance of wrinkles? What is the common name of the infection characterized by fever, sore throat, and enlarged lymph nodes that is caused by the Streptococcus bacteria? Which supplement is presumed to influence heart health due to its concentr ation of omega 3 fatty acids? Philanthropy What term describes the active effort to enhance human potential and promote public welfare, either through deeds or charitable donations? Which actor won an Oscar for his role as a finicky, obsessive compulsive novelist in As Good as it Gets ? Which Irish lead singer of U2 is now known for his humanitarianis m as well as his music? An electroencephal ogram is used to measure electrical activity in which organ? Photosynthesis What do you call the chemical process p erformed by plants, algae, and bacteria by which carbon dioxide is converted into food through the use of energy from the sun, releasing oxygen as a byproduct? In what famous Dutch painting does a pearl earring serve as the focal point of the work? What d o you call a material, such as chlorophyll, that changes color by selectively absorbing certain wavelengths of light, and in humans is found in the skin? A Bloody Mary is a drink made of tomato juice and which liquor? Planetarium What is the name of the b uilding where one can view projected images of celestial bodies on the inner surface of a dome? What type of engineer is responsible for planning, designing, constructing, maintaining, and operating infrastructures while safeguarding the interests of socie ty and the environment? The invention of which instrument was a necessary requirement for the development of astronomy into a modern science? What is the process that determines which biological traits survive in a population as a function of reproductive advantage?
158 Poinsettia What is the name of the flower indigenous to Central America that has large bright red leaves and is especially trendy at Christmas? In one of the most poignant movie moments of time, Rhett Butler bids farewell to Scarlett O'Hara by saying 'Frankly my dear, I which 1939 film? At Christmas time, what activity are two people supposed to do if they find themselves under a mistletoe? Which Aerosmith rocker and reality show host became a grandfather in 2004? Reinca rnation What do you call the spiritual belief that the soul of a person is born again in another body after death? The bank is likely to repossess a house if which type of payment is not made after several months? The oldest living denomination, Hinduism's basic tenets include a belief in a cyclical pattern of spiritual birth that is determined by a person's _______, where beneficial consequences are the direct product of past moral actions. The successful passage of food from esophagus to the stomach begi ns with which activity that is initiated in the mouth? Rosary In the Catholic faith, what do you call the string of beads that is used to count a series of prayers that make up a devotion? What unit of time is determined by the rate at which the Earth rot ates on its axis? What do you call a devotional offering of food, objects, or the lives of animals or people, usually left on an altar? How many miles do participants complete in a marathon, a commemoration of the fabled distance that the Greek messenger s oldier traveled to Athens to announce victory after the Battle of Marathon?
159 Simile What is the figure of speech that is used to compare one thing to another, unlike thing by the use of 'like' or 'as' ? Simmering is a food preparation technique that invo lves cooking food in hot liquid kept at or just below which point? Which popular hit from the 1980s uses a metaphor to liken the experience of love to war? Which website debuted in 1997, and is now the most visited website on the Internet? Sovereignty Wh at is the state of having absolute governing authority over a geographic area? At many weddings, the husband and wife make the solemn vow to love one another for 'better for worse', for 'richer for poorer', and in which other opposing conditions? On what day do Americans celebrate their independence from Britain, the same day that the Founding Fathers signed the Declaration? What is the little boy's name in the classic tale about magic beans that grow into a giant beanstalk? Suspenders What do you call th e fabric or leather straps worn over the shoulders to hold up trousers? What is the most commonly diagnosed behavioral disorder in children, characterized by an inability to sustain attention, hyperactivity, and/or impulsivity? Because of its utility as a protective garment that covers the torso and legs, overalls were required for workers manufacturing which reactive material? What is the world's best known and most prestigious bicycle race? Toga What is the term for a loose outer garment worn by citizen s of ancient Rome? Which best selling home exercise system promises unrivaled results in 90 days through the use of total body cross training that prevents the body from adapting to the workout? What country is associated with the sari, a piece of cloth wo rn by women that is usually wrapped around the waist, with one end then draped over the shoulder baring the midriff? A clever strategy in the game of hangman is to start by guessing the most frequently occurring letters, which include all the letters of wh ich type?
160 Torpedo What do you call a large, cigar shaped, self propelled projectile that is launched above or below the water's surface? In which Shakespearean play does the lead female character become so tormented with guilt that she tries to wash imag inary blood from her hands? What kind of radiation is often used to guide a missile to its destination using the thermal (heat) energy emitted by an object? What kind of funding is provided by the government to help cover college expenses on the basis of need? Tyrant What do you call an oppressive and cruel ruler (not dictator) that is unrestrained by the law and puts his or her own interests above the interests of the population? The Tidy Towns Awards were started in which country (that is also a contin ent) in order to reduce littering and bolster civic pride? What is the name of the 'Red Emperor' or 'Chairman', who was the leader of the Chinese Revolution in 1949 and established the country now known as the People's Republic of China? In the U.S. army, what do you call the person responsible for the instruction of new recruits about the customs and practices of military life, and is especially important during Basic Training? Venison What do you call the edible meat of a game animal, most commonly deer ? It is important to ventilate the air in occupied buildings in order to limit the amount of which chemical compound that can be toxic at high concentrations? Buffalo meat has increased in popularity among health enthusiasts because it has significantly l ess ____ relative to beef. What do you call the form of transportation, usually a boat, used to carry passengers and sometimes automobiles and cargo across a body of water? Vernacular What do you call a nonstandard language or dialect (not slang), such as ebonics, that is specific to an area or population and is used in place of a more established foreign language? Which NFL wide receiver plays for the Tennessee Titans, holds the record for single season touchdown receptions, and has one of the highest rec orded vertical leaps in sports? What colloquialism is commonly used throughout the South as a contraction of 'you' and 'all'? According to a theory of developmental psychology, children raised by an authoritarian parent are likely to have _________ self es teem.
161 APPENDIX B PROPER NAME TARGET, PRIME, AND UNRELATED QUESTIONS Proper Name Target Target Question Phonological Prime Question Semantic Prime Question Unrelated Question Alan Greenspan Which U.S. economist served as Chairman of the Federal Reser ve from 1987 to 2006? Since 1984, Alex Trebek has been the host of which trivia game show? Donald Rumsfeld was Director of the Office of Economic Recovery from 1969 1971, before holding which Cabinet office under two different presidents? In 1998, Sammy S osa was in pursuit of breaking the Major League Baseball record for the number of _____. Alexander Hamilton Which founding father of the United States was the first Secretary of the Treasury and was killed in a duel with political opponent Aaron Burr in 1 804? Which TV sitcom currently stars Alec Baldwin as a TV network mogul working at a famous New York address? John Adams was a Founding Father, Federalist, lawyer, and diplomat before becoming the _________ president of the U.S.. In one of his most popula r songs, singer songwriter Bob Dylan advises that the answer is 'Blowin' in the ______'. Alfred Hitchcock Which English filmmaker was the master of suspense, making films such as Psycho and Vertigo ? Which famous equation relevant to mass and energy did Al bert Einstein discover? What Stephen King novel about an author nursed back to health by an obsessed fan became a movie thriller in which Kathy Bates won a Best Actress Oscar? Celebrity chef Bobby Flay has a show on the Food Network that focuses on using what cooking method?
162 Amelia Earhart Who was the first female aviator to fly solo across the Atlantic Ocean, later disappearing in 1937 after attempting a circumnavigation al flight of the globe? Twenty six year old star Amanda Bynes got her big break work ing for which children's TV network? Molly Brown was nicknamed the 'Unsinkable' after her survival and heroism during the sinking of which famous ship? Lucy Liu costarred in which 2000 film adaptation of a popular 1970s TV series about sexy female private investigators? Andy Griffith Which actor had a self titled TV comedy show in the 1960s and then played the lead character in the legal drama Matlock in the 1980s and 90s? Journalist Anderson Cooper is the host of the news show AC 360 which is shown on w hat cable channel? Actor Leslie Nielsen became known for his spoof films, including the popular comedy movie series, The ____ Gun In which Nebraska city is Warren Buffett's company Berkshire Hathaway located, the largest city in the state that resides on the Missouri River? Andy Warhol Which controversial artist of the 1950's 1980's was a leading figure in the 'pop art' movement, best known for his paintings of iconic American products and celebrities? Aside from his Travel Channel show that documents hi s exotic travels and culinary quests, Anthony Bordain is also a frequent guest judge on Bravo's Top _________. Known for his unique drip style paintings, Jackson Pollock was given a memorial exhibition in what New York City museum the same year as his deat h? English writer Rudyard Kipling is best known for penning which book of fables about Mowgli's adventures with exotic creatures in the Indian jungle? Annie Oakley What is the stage name of the sharp shooting female star in Buffalo Bill's Wild West show, best known for her tricks with a .22 caliber rifle? Annabel Lee is the last completed poem by which three named American author known for his gothic, macabre style? Calamity Jane was a frontierswoman and scout in the late 19th Century known for fighting w hat indigenous group in the American West? American author Emily Post was famous for writing about which topic, a guide for social behavior?
163 Anthony Hopkins Which actor is best known for his portrayal of Hannibal Lector in the movie Silence of the Lambs ? Photographer Ansel Adams is best known for his black and white photographs of the American West, especially which national park? Starring in seven films as this character, what is Sean Connery's most famous role? American football coach Vince Lombardi is best known for winning the first two Super Bowls coaching which NFL team? Arthur Miller Who wrote the classics Death of a Salesman and The Crucible and was once married to Marilyn Monroe? The combination of which two beverages creates an Arnold Palmer ', a drink named after the famous golfer? Neil Simon's plays were made famous on what world renowned New York theater district? Canadian actor Ryan Reynolds was once married to which curvaceous Lost in Translation actress? Atticus Finch Which literary cha racter was the heroic protagonist lawyer in the Pulitzer Prize winning novel To Kill a Mockingbird ? Adam Smith the Father of Economics, developed which economic principle that espouses private ownership, competitive markets, and a government free market s ystem? Holden Caulfield was the narrator of what 1951 J.D. Salinger novel about a teenager coming of age, struggling with his sexuality, and battling mental health? A famous explorer, Marco Polo is also the name of a game of tag played in a _____. Audrey Hepburn Who was a Belgian ballet dancer before starring as the lead actress in Breakfast at Tiffany's and the movie adaptation of the musical My Fair Lady ? Singer Aubrey O'Day the MTV reality star, was also a contestant on which TV show that is famous fo r the catch phrase 'You're Fired!'? Before her untimely death at age 43, Natalie Wood was a successful Hollywood actress, starring in which movie musical about warring gangs in New York City? Athlete and model Gabrielle Reece is known for playing which pr ofessional sport on the sand?
164 Austin Powers Who is the title character in the comedic film series of the late 1990s about a British spy's heroic quest to bring the villain Doctor Evil to justice? Oscar De La Hoya won 10 world titles in six different cl asses competing in which sport? Clark Griswold is the fictional patriarch in the comedy series about a family continuously experiencing minor disasters while trying to enjoy what type of annual event? Bruce Jenner is a former track and field athlete and g old medalist who is now better known for his role in the reality TV show Keeping up with the __________. Barney Rubble In the animated TV show The Flintstones who is the main character's best friend? Barnaby Jones was a 1970s TV series about a private de tective firm owned by a father and his __________ in law. What is the name of Fred Flintstone' s wife? On Saturday Night Live Will Ferrell became famous for his impersonation of which U.S. President? Benedict Arnold Which American Revolutionary War gene ral's name is now synonymous with treason due to the fact that he secretly plotted to surrender West Point to the British and eventually changed sides after getting caught? Bennie and the Jets was a 1974 hit about a fictional band that was performed by wh ich iconic British singer song writer? Paul Revere was a famous patriot of the American Revolution, most famous for alerting the colonial militia of what approaching danger? Johnny Depp found major box office success playing the role of Captain Jack Sparro w in which movie series based on a ride at Disney theme parks? Bette Davis Which actress was the first person to accrue 10 Academy Award nominations, including the movies All About Eve What Ever Happened to Baby Jane and Jezebel ? Beverly Cleary is an Am erican children's author, whose best known characters include the Quimby sisters, Beezus and _________. Joan Crawford was a Hollywood star who played the lead in Mildred Pierce and co starred in What Ever Happened to Baby Jane before her reputation was ta rnished by a tell all book written by which member of her family? Rap and R&B artist Faith Evans is known as the widow of which New York rapper, famously killed in a drive by shooting in Los Angeles?
165 Betty Boop Which buxom cartoon character first appear ed in 1930 and is considered the first and most famous sex symbol in the world of animation? Despite the lack of historical evidence, Betsy Ross is widely credited as having made the first American _________. Olive Oyl was the cute cartoon love interest o f Popeye the Sailor Man, who advocated eating which food item to stay strong? Actress Amy Smart currently plays a recurring role on which Showtime series about a dysfunctional family in Southside Chicago? Beyonc Knowles What is the name of the singer who was a lead member of the musical group, Destiny's Child, starred in the film Dreamgirls and is married to the rapper, Jay Z? English children's author Beatrix Potter is most recognized for which story about a mischievous young rabbit in a garden? Mary J Blige is the only artist to have won which type of award in the categories of Pop, Rap, Gospel, and R&B? Actress and sex symbol Jessica Alba has how many children with her husband Cash Warren? Calvin Klein Which American fashion brand known for its mini malist style in sportswear, underwear, and fragrance, has also been made famous by the overt sexuality of its print ads, including a controversial feature of 15 year old Brooke Shields? The 'Iron Man' Cal Ripken was inducted into which sport's Hall of Fam e in 2007? Tommy Hilfiger is an American fashion brand that gained the interest of artists from which music genre, who popularized wearing the apparel several sizes too large? Singer Josh Groban performed 'The Prayer as a duet for the ending ceremonies of the 2002 Winter Olympics, hosted in which Utah city? Carmen Electra What is the stage name of the actress/ model who gained fame for her appearances in Playboy and Baywatch, for her highly public relationships with athletes and rock stars, and for her po pular 'striptease' workout videos? What was the first name of Carmela Soprano's husband on HBO's The Soprano's the show's lead character? Baywatch actress Pamela Anderson was previously married to Tommy Lee, who is the drummer of which heavy metal band? What is the name of the first book that became a bestselling series of novels written by Diana Gabaldon ?
166 Christopher Reeve Which actor known for playing Superman in multiple movies in the late 1970s and 80s became a quadriplegic after being thrown from a horse? What is the American name of the sport that Cristiano Ronaldo plays for the Portuguese national team as well as professionally for a club in Madrid? Actor Michael Keaton is best known for playing the roles in two Tim Burton films, the title role in Beetlejuice and Bruce Wayne in what superhero flick? Serial killer Jeffrey Dahmer made his murders particularly gruesome by eating some of the flesh of his victims, a practice known as _____. Condaleeza Rice Who was the first female African American s ecretary of state serving under George W. Bush for his second term? American news anchor and reporter Connie Chung has been married to which TV talk show host since 1984? Before joining the Obama administration, Hillary Clinton served in which upper house of the United States legislative branch? Actress Linda Blair is best known for playing the possessed child Regan in which 1973 horror film? Davy Crockett Which American folk hero died at the Battle of the Alamo and is referred to in popular culture as th e 'King of the Wild Frontier'? Dale Earnhardt was a dominant force in which sport prior to his death in 2001? Paul Bunyan a giant lumberjack of North American folklore, is often depicted with what type of animal companion? Winning 14 Grand Slam titles, P ete Sampras was once ranked #1 in the world before retiring from what sport? Demi Moore What is the name of the actress whose film credits include St. Elmo's Fire Ghost and Indecent Proposal and who recently ended her marriage to the much younger Ashto n Kutcher? Debbie Harry pop punk star of the early 1980s, was the front woman of which band whose name reflects the lead singer's platinum hair? Ally Sheedy is best known for her roles in 1980s brat pack films, including St. Elmo's Fire and which film abo ut a group of high school students serving a punishment on the weekend? Eva Peron the First Lady of Argentina from 1946 to 1952, was a powerful spiritual and political leader in the country, whose life inspired which musical?
167 Diane Sawyer What is the n ame of the current anchor of ABC World News who previously held the position of co anchor for ABC's Good Morning America ? Diana Ross's first number 1 hit was the motown favorite 'Ain't No ______ High Enough'. Journalist Katie Couric has previously anchored CBS Evening News as well as which morning news/talk show? Before acting on Broadway, Cathy Rigby won a silver medal in which event at the 1970 Gymnastics World Championships? Dolly Parton Which American singer songwriter is one of the most successful fe male country artists of all time and is famous for her role in movies like Steel Magnolias and Straight Talk ? Dominique Dawes was a member of the Magnificent Seven in 1996, a team that brought the United States its first ever Olympic team gold medal in whi ch sport? Emmylou Harris is a 12 time Grammy winning folk and country musician who has collaborated with numerous other artists, including Willie Nelson and which 'Heart of Gold' singer? Seattle native Amanda Knox served four years in jail after being cha rged with murder when she was studying abroad in which European country? Donald Trump Which American business magnate gained public notoriety in the 2000s after launching the reality TV show Celebrity Apprentice and through becoming increasingly involved with American politics, most recently in the Republican primaries? Currently a free agent, Syracuse grad Donovan McNabb plays which football position? Millionaire Bill Gates is the former CEO and current chairman of which computer software company? Twent y five year old Shaun White is a two time Olympic gold medalist in which winter sport?
168 Eddie Murphy Which comedian went on to acting fame in the movie Beverly Hills Cop in the 1980s, The Nutty Professor in the 1990s, and most recently in Shrek as the voi ce of Donkey? Artist Edgar Degas perhaps best known for his paintings of dancers, was born in which French city? African American actor and comedian Martin Lawrence is widely known for playing which 'large' female character in a movie series? The 17 year old pop sensation Justin Bieber was discovered after posting some of his videos on which video sharing website? Eli Manning Who is the quarterback for the NY Giants, leading them to win the 2012 Super Bowl, and who has a brother who is the former quarter back of the Indianapolis Colts? Actor Elijah Wood is best known for his portrayal of Frodo Baggins in which fantasy adventure trilogy? At the age of 23, Ben Roethlisberger became the youngest Super Bowl winning quarterback playing for which NFL team? Nico las Cage won the Academy Award for his performance as a suicidal alcoholic in which 1995 film based on a novel by the same name? Elizabeth Taylor Which iconic film actress, known for her lavish lifestyle and long term love affair with Richard Burton, star red in Cleopatra and Suddenly, Last Summer and recently died in 2011? Elaine Benes was the friend and former girlfriend of the main character on which NBC sitcom that also featured George Costanza and Cosmo Kramer? Feisty Irish actress Maureen O'Hara sta rred opposite which famously rugged American movie star in The Quiet Man ? American pop singer Colbie Caillat won a Grammy in 2010 for the song Lucky, a duet she performed with which male singer songwriter? Ernest Hemingway Who is the author of A Farewell to Arms and The Sun Also Rises ? Gator football coach Urban Meyer won how many national titles in his tenure from 2005 to 2010? What was the name of Mark Twain's most esteemed novel, often referred to as the Great American Novel? After the success of his in itial feature film, director Spike Lee was offered a job directing commercials for which sports company?
169 Eva Longoria What is the name of the Latina actress that plays the character of Gabrielle Solis on the television series Desperate Housewives and wa s once married to NBA star Tony Parker? Yves Saint Laurent is a luxury fashion brand that goes by which self titled three letter label? Actress Salma Hayek was born in which Latin American country? British singer Susan Boyle received international attenti on singing 'I Dreamed a Dream on which reality competition show? Fidel Castro A former hero of the Cuban Revolution, who was the highly controversial communist president of Cuba from 1976 to 2008? In 2005, Philip Seymour Hoffman won the Best Actor Academ y Award for which biographical film about an American author? Marxist revolutionary Che Guevara began forming his socialist ideology on a 9 month journey through South America on what type of vehicle? Tom Wilkinson won a Golden Globe for his portrayal of B enjamin Franklin in the popular miniseries about the Founding Fathers on which premium TV network? Franklin D. Roosevelt Highly regarded as one of the best presidents in United States history, which man was elected to four terms in office and navigated t he country through the Great Depression and most of World War II? Frankie Valli and the Four Seasons is one of the best selling vocal groups of all time, reaching peak fame in the 1960s with hits like 'Sherry' and 'Big Girls Don't _____________ ____'. Harr y Truman became the 33rd president at the advent of the Cold War and created the Truman Doctrine to contain which spreading political movement in Europe? What is the English title of Victor Hugo's novel about Quasimodo and Esmeralda? Freddy Krueger What is the name of the fictional villain in the A Nightmare on Elm Street film series, a disfigured monster who kills victims in their dreams with a glove armed with razors? Former slave turned statesman Frederick Douglass was a leader in the abolitionist move ment during the Civil War and gave the keynote address at which president's funeral service? In the 1988 horror film Child's Play the main villain Chucky is not a person but rather which type of toy? Early on, American artist and illustrator Norman Rockwe ll was the art director for Boys' Life the publication produced by which major youth organization?
170 Gerald Ford Who was the only US president never elected as president or vice president by the electoral college, and became the 38th president after the re signation of Richard Nixon? English actor Jeremy Irons starred opposite Helen Mirren in the historical miniseries about the reign of which queen? The 40th president of the United States, Ronald Reagan was a darling of the Republican party due to his policy of supply side economics, now known by which nickname? Richard Pryor was regarded as one of the most influential stand up comedians of his era, whose comedy acts often confronted what serious social issue in America? Harley Davidson What is the name of the American motorcycle company that is known for its emphasis on freeway cruising and motorcycle customization? In 1977, Harvey Milk became the first openly gay person be elected to public office in what U.S. state? Luxury automobile manufacturer Mercedes Benz has its roots in which European country? In the 1980s, Willard Scott was the weatherman for The Today Show and began the practice of sending birthday wishes to people starting at which age? Harrison Ford Which actor is best known for his performanc es as Indiana Jones, as Dr. Richard Kimble in The Fugitive and as a pilot in the original Star Wars trilogy? One of the most famous escapes demonstrated by magician Harry Houdini involved him being suspended upside down in a locked glass and steel cabinet full of _____. One of Tommy Lee Jones' most notable roles was a federal marshal in which movie about an on the run doctor wrongfully accused of murder? In 2009, baseball player Alex Rodriguez admitted to using _____ from 2001 to 2003.
171 Herbert Hoover W hich president tried to combat the Great Depression by increasing corporate taxes and initiating massive public works projects, such as a dam on the border of Arizona and Nevada that is named after him? Author Herman Melville is best known for writing whic h Great American Novel about Ishmael, Captain Ahab, and a white whale? William Taft was the only U.S. president to also be appointed to which highest office in the federal judicial branch? Michael Crichton was an author, producer, and screenwriter most oft en associated with the Sci Fi genre, including which book turned movie about a theme park dedicated to genetically recreated dinosaurs? Herman Cain Which African American Tea Party activist sought the 2012 Republican Party presidential nomination but wit hdrew after repeated allegations of sexual misconduct? Herbie Hancock is an American musician of which music genre that features blue notes, improvisation and syncopation? African American civil rights activist Al Sharpton was a candidate for the 2004 Demo cratic presidential nomination, and now makes regular appearances on which cable news channel with a reputation for promoting conservative political positions? What nickname was bestowed upon Reggie Jackson due to his clutch hitting in the postseason play ing for the New York Yankees? Hugo Chavez Who is the current president of Venezuela whose platform is known as the 'Bolivarian Revolution' and has been implementing socialist reforms in the country since 1999? The band Huey Lewis and the News contributed to the soundtrack of which 1985 film about a time traveling teen? Controversial socialist politician Daniel Ortega is the current president of which Central American country, located north of Costa Rica? David Letterman recently surpassed which television icon for hosting the longest running late night TV show in America?
172 Isaac Newton One of the most influential scientists in history, who is known for developing the three laws of motion and is fabled to have discovered gravity when an apple hit him on th e head? Isaiah Mustafa became famous in 2010 after starring as the hunky 'Man Your Man Could Smell Like' in commercials for which deodorant brand? French philosopher and mathematician Ren Descartes was a key figure in the ________ Revolution, the historic al era characterized by an expansion in knowledge of physics, astronomy, biology, medicine, and chemistry. The most famous song by Woody Guthrie is 'This _____ is Your Land'. Jackie Robinson Who was the first African American man to play Major League Bas eball in the modern era, joining the Brooklyn Dodgers in 1947 and bringing an end to racial segregation in professional baseball? Singer songwriter Jackson Browne wrote many of the hits for which American band, best known for the songs Hotel California and Take it Easy ? Willie Mays began his professional baseball career in the 'Negro Leagues,' before getting picked up by the New York _______, a franchise now in San Francisco. Fearing that his creativity would be stifled by new standards passed by the Federa l Communications Commission, outspoken host Howard Stern moved his famously crude radio show to which satellite radio company in 2006? Janet Reno Who was the first female Attorney General of the U.S., nominated by then President Bill Clinton? Nicknamed t he Queen of Rock and Roll, Janis Joplin was one of the biggest attractions at which legendary 1969 music festival held at a dairy farm in upstate New York? Madeleine Albrigh t was the first woman to be appointed to which position in the United States Cabine t, the office primarily concerned with foreign affairs and diplomacy? Actress Kristen Davis is most famous for playing the role of Charlotte on which comedy drama series about a group of four women living in New York City?
173 Jefferson Davis Who was presid ent of the Confederacy during the Civil War, criticized by historians as an ineffective wartime leader? Motorcycle customizer and TV personality Jesse James created a huge media scandal in 2010 when he was caught having multiple affairs during his marriage to which famous actress? Confederate commander Robert E. Lee first established himself as an exceptional tactician and officer in which war between neighboring countries? Before being fired in 2011, Charlie Sheen was the highest paid actor in television working on which show? Jerry Springer Who was the host of the self titled TV show about dysfunctional families and relationships, which frequently resulted in fights between guests? In which HBO TV series about a movie star and his friends from Queens doe s Jeremy Piven play a Hollywood agent? Phil Donahue's talk show had a 26 year run on national TV and often focused on social issues that divided which 2 political factions? Notorious Wild West train robber, bank robber, and gang leader Butch Cassidy was f orced to flee the country with his accomplice, who was known by what nickname? Julia Child What is the name of the American chef who introduced French cuisine through her cookbook Mastering the Art of French Cooking and her television show The French Che f ? American actress and vaudevillian Judy Garland is perhaps best known for playing a Kansas farm girl named Dorothy in which 1939 film? Martha Stewart who gained fame through the success of her TV show which focused on entertaining, decorating, and cooki ng, also had a line of home dcor at which discount store? Hollywood's top female legend Katharine Hepburn holds the record for winning how many Academy Awards for Best Actress? Julie Andrews Which singing actress is best known for her lead roles in Mary Poppins and The Sound of Music ? Author Judy Blume has written many novels, such as Are You There God? It's Me, Margaret for which age group? Barbra Streisand who began her acting career in movie musicals such as Funny Girl and Hello, Dolly! is widely k nown to suffer from extreme _______ during live performances. Rachael Ray hosts a cooking show on the Food Network based on the premise that a delicious meal can be prepared in how many minutes?
174 Keanu Reeves What Canadian actor is best known for his role s in Point Break Speed and The Matrix series? Comedian Kenan Thompson got his start as a teen star on Nickelodeon, and was a cast member on which NBC sketch comedy show that airs every weekend? Val Kilmer rose to fame in the mid 1980s starring in comedie s and then blockbuster action films, including which film about young Naval aviators in Fighter Weapons School? Larry Birkhead is known for having a child with which former Playmate who died of an overdose of prescription drugs in 2007? Laura Bush What f ormer Texas librarian is married to the 43rd president of the United States and is the mother to twin daughters? Actress and model Lauren Bacall known for her husky voice and sultry looks, was married to which Casablanca leading man? Elizabeth Dole whos e husband ran for president in 1996, represented which southern state in the U.S. Senate? One of the most successful models of the 1990s, Cindy Crawford's trademark is a small ________ above her lip. Leonardo da Vinci The original 'Renaissance Man,' which Italian artist painted the Mona Lisa and The Last Supper, and is also revered for his design ingenuity? According to multiple government agencies, Lee Harvey Oswald assassinated which U.S. president in Dallas, Texas in 1963? On which chapel's wall did Re naissance artist Michelangelo paint one of the most famous frescos in Western history? Paul Michael Glaser is best known for playing one of the protagonists in which 1970s cop thriller that ran for four seasons? Louis Armstrong Who was the jazz musician a nd singer famous for his trumpet playing and scat singing, as well as his hits such as 'What a Wonderful World' and Dream a Little Dream of Me '? Lewis Carroll is most famous for writing which children's story about a girl who falls down a rabbit hole into a fantastical land ? Because of his exceptional writing and improvisation, jazz legend John Coltrane was awarded which prize in 2007, typically associated with journalists and writers? In which movie did Tom Hanks say the quote 'Mama always said life is like a box of chocolates: You never know what you're gonna get'?
175 Lucille Ball Which red headed comedienne had one of Hollywood's longest careers and reached peak fame in a 1950s sitcom about a zany housewife that co starred her then husband Desi Arnaz? Which sister is considered the 'beauty' of the family in Louisa May Alcott's best known novel about Jo, Meg, Beth, and Amy March? What was the name of Carol Burnette's popular variety show that ran from 1967 1978, that included comedy sketches, song, and d ance? University of Florida grad Erin Andrews is a co host on which ESPN college preview show? Mahatma Ghandi Who was the prominent leader in the struggle for Indian independence in the first half of the 20th century, who was a staunch advocate for non v iolence? Before a fight, boxer Mohammed Ali famously told reporters that he was going to 'float like a butterfly, and sting like a _______'. The Dalai Lama is the title given to the highest 'lama' or teacher in the Tibetan branch of which Eastern based rel igion? English philosopher Francis Bacon popularized the use of which 'method' that stresses the importance of forming falsifiable hypotheses and collecting evidence through experiments and observation? Maria Shriver Which journalist came from a highly f amous American family and was the wife of the 'Governator', serving as First Lady of California from 2003 2011? In 1992, Marisa Tomei won the best supporting actress Oscar for her performance in the comedy, My Cousin _________ Author and attorney Caroline Kennedy the last surviving child of John and Jackie, endorsed which nominee in the 2008 democratic primary? In 2008, People magazine reported that Christina Applegate had been diagnosed with what disease? Marie Curie What is the name of the female scie ntist who laid the foundation for the study of radioactivity and won Nobel prizes in physics and chemistry? Rock singer Melissa Etheridge is known for her raspy vocals and her advocacy of _________ rights. Known as the 'lady with the lamp,' Florence Nighti ngale tended to wounded British soldiers during the Crimean war and laid the foundation for which health profession? In which futuristic film series did Linda Hamilton play the role of Sarah Connor?
176 Marlon Brando Which award winning actor is known for hi s performances in A Streetcar Named Desire Apocalypse Now and his portrayal of Vito Corleone in The Godfather ? In the 16th century, Martin Luther was a key figure in the Protestant Reformation, forming what sect of Western Christianity? Iconic bad boy Ja mes Dean cemented his legendary status in American movie history, acting in only three films before dying of which cause at age 24? In his famous 1995 trial, O.J. Simpson was acquitted of killing his ex wife and her friend after he was famously unable to p ut on what type of garment found at the crime scene? Marty McFly What was the name of the character played by Michael J. Fox in the Back to the Future trilogy? Singer Marvin Gaye's song 'I Heard It Through the Grapevine' became the signature tune of which clay animated singing group in the 1980s? In the 1986 comedy film Ferris Bueller's Day Off Ferris and his friends skip school to spend the day having adventures in what U.S. city? David Schwimmer is best known for playing Ross on what 10 season sitcom a bout a group of young people living in New York City? Mary Magdalene Which controversial biblical figure is regarded as one of the most important female disciples but is referred to as a prostitute in some texts? In 1962, American sex symbol Marilyn Monro e famously serenaded a U.S. president with what song at a celebration of his 45th birthday? Joan of Arc is considered a heroine of the Catholic Church after leading which army to several important victories in the 100 Years War before being captured and bu rned at the stake? In 2009, Vanessa Redgrave's daughter Natasha Richardson died after a head injury sustained during which activity? Michael Dukakis Which Democratic presidential nominee of Greek heritage ran against George H. W. Bush in 1988 and lost? O ne of the most influential musicians of the 20th century, Miles Davis was a jazz music pioneer and was best known for his prowess with what horn instrument? Businessman Ross Perot ran in the 1992 presidential elections as an unusually successful __________ candidate, and then in 1996 as a Reform Party candidate? In the fantasy novel series about young wizards coming of age, what is the name of the school that Harry Potter and his friends attend?
177 Michael Phelps Which American Olympic Swimmer won 16 gold me dals combined in the 2004 and 2008 Olympics but received negative fame when pictures of him holding a bong appeared on the internet? Mighty Mouse was an animated mouse superhero created by which major movie studio, also known for the popular animated serie s The Simpsons ? American swimmer and Olympic medalist Ryan Lochte specializes in the individual medley and what kind of swimming stroke? Which neighborhood in New York City has been a dominant African American residential, cultural and business center sinc e the 1920s, and is often cited as the inspiration for Langston Hughes poetry? Mikhail Gorbachev Who was the last General Secretary of the Communist Party of the Soviet Union whose attempts at reform and reorientation of Soviet strategic aims contribute d to the end of the Cold War? Actor Mickey Rooney did the voice of which 'jolly, old elf' character in four different Christmas TV animated specials? Boris Yeltsin became the first president of the Russian Federation in 1991 after the formal dissolution o f which Single Party Communist State? Which 1984 film stars Kevin Bacon as a teen who moves to a small town where dancing and rock music have been banned? Monica Lewinsky What is the name of the woman involved in an affair with President Bill Clinton in 1 995 and 1996 that caused his impeachment? Molly Ringwald was a member of the 'brat pack' of teen stars in the 1980s films, whose film credits include Pretty in Pink and ________ Candles Paula Jones who sued Bill Clinton for sexual harassment prior to his impeachment, was an employee of which U.S. state where Clinton had previously been governor? Now a two time Grammy winner, Kelly Clarkson got her first break after winning the inaugural season of which reality competition show? Monty Python What is the name of the British comedy troupe that created a sketch comedy show called _______ _______ 's Flying Circus which later developed into films, books, albums, and stage productions? After his diagnosis in the late 1990s, talk show host Montel Williams is no w actively involved in a nonprofit foundation dedicated to which disease? The comedy duo that consisted of Cheech Marin and his partner Chong gained success in the 1970s and 80s with an act that focused on their love of which drug? Canadian rock singer Br yan Adams had a huge hit with which song about a certain season in the late 1960s?
178 Nathaniel Hawthorne Which American writer composed the classics The Scarlet Letter and The House of Seven Gables ? One of the most successful military commanders in history, the French ruler Napoleon Bonaparte is cited as having an inferiority complex because of what physical 'flaw'? American essayist Ralph Waldo Emerson gave a now famous speech entitled the 'American Scholar' to which academic honor society, one of the olde st in the United States? French film actor Olivier Martinez recently announced his engagement to which Catwoman actress, whom he has been dating since 2010? Orville Redenbach er Which American businessman started a self titled popcorn brand in 1970? Orson Welles was best known for his Broadway production Caesar adapted from which best known English playwright's work? H.J. Heinz founded a food manufacturing company in the 1860s, which is now best known for producing which product? On what fictional island do Peter Pan Tinker Bell, and the Lost Boys live? Peter Parker Which fictional teenage superhero is the alter ego of Spider Man in the Marvel comic book series? P.T. Barnum was an American showman and businessman, associated with what type of travelling entertainment troupe, featuring clowns, acrobats, and animals? Clark Kent is the secret civilian identity of which DC Comics superhero? Will Smith starred in which 1997 sci fi comedy about a secret agency dedicated to keeping Earth a 'neutral zone' for ex traterrestrial aliens? Rosa Parks What African American hero of the Civil Rights Movement refused to give up her bus seat in 1955, and whose arrest sparked the Montgomery Bus Boycott? After hosting her own talk show for 6 seasons, in 2006 Rosie O'Donnell joined the cast of which ABC daytime show featuring a council of diverse women? Civil rights leader Coretta Scott King initiated the movement to make the anniversary of her husband's assassination a national holiday on the 15th of which month? Despite the lack of historical evidence, an oft repeated legend holds that when Marie Antoinette was told the peasants had no bread to eat, she answered with what famous quote?
179 Rudy Giuliani What is the name of the former New York City mayor who was a supportive lea der during the 9/11 attacks? In 2011, Rupert Murdoch faced allegations that his companies, including News of the World and other tabloids, had been using which tactic to illegally acquire information about celebrities and politicians? The 12th richest man in the United States, Michael Bloomberg is currently the mayor of which city? Conrad Hilton was an American business man and inventor best noted for starting which chain? Salvador Dal What Spanish surrealist painter of the 20th century has a museum dedi cated to his work in St. Petersburg, Florida, and whose paintings were noted for their extensive use of symbolism? Actor Sal Mineo is best known for his performance as John 'Plato' Crawford in A Rebel without a _______ One of the most influential artists of the 20th century, Pablo Picasso split his time between Paris and which Spanish city during the early part of his career? Before his independent career, Justin Timberlake was a member of which popular boy band of the late 1990s and early 2000s? Sherloc k Holmes What fictional detective created by Arthur Conan Doyle is known for his logical reasoning and is often aided by his sidekick Dr. John Watson? Sherman Hemsley played George Jefferson on both The Jeffersons and its parent sitcom, All in the _______ The comic strip Dick Tracy featured a quick witted and fast shooting detective who was depicted wearing a hat and trench coat of which color? Dan Aykroyd is a founder of which blues and soul revivalist band that was created as part of a comedy sketch on Saturday Night Live ? Sigmund Freud Who is the father of psychoanalysis, known for his controversial theories about repression, the unconscious mind, dream analyses, and the Oedipus Complex? Siegfried Fischbacher and his partner Roy had one of the most vi sited shows in Las Vegas involving daring tricks with lions and tigers, which were which exotic color? Ivan Pavlov was a famous Russian psychologist who accidentally discovered classical conditioning while researching the salivary gland in which species? M eaning 'let do' or 'leave it alone', what is the term for an economic policy based on markets without government intervention, an ideal advocated by small government conservatives such as Calvin Coolidge ?
180 Tony Bennett Which Italian American crooner has e njoyed a prolific career crossing pop, classics, show tunes, and jazz and whose signature song is 'I Left My Heart in San Francisco'? Actor Tobey Maguire starred in what 2003 film about an overlooked racing horse who had unexpected success? Crooner Frank Sinatra is famous for singing about 'New York, New York', a city that never _______. On which two South Eastern Conference (SEC) football teams has Steve Spurrier acted as head coach? Tony Hawk Which retired skateboarder was a pioneer athlete in the sport made famous by completing the first '900' and for the video game series bearing his name? American country singer Toby Keith had a long lasting number one hit with the song '__________ for My Horses' Before his tragic death in 2010, Andy Irons was a wor ld class professional surfer, inspiring a line of surfboards produced by which iconic surf brand? Matthew Fox is best known for his role as Jack on which supernatural drama series that started with a plane crash and ended in 2010? Tyler Perry Which Afric an American actor, director, and producer was the highest paid man in entertainment in 2011, largely due to the success of the Madea series? In which sport has Tiger Woods won 14 professional major championships and was the youngest man to achieve a career Grand Slam? African American comedian Bernie Mac had his most notable film role in which movie about a band of 11 criminals attempting to pull off a heist on a Las Vegas casino? Philosopher David Hume is known for empiricism, which relies on the idea tha t the mind is a _________ _______, or tabula rasa. Virginia Woolf Which feminist author of Mrs. Dalloway is praised for her inclusion of stream of consciousness and psychologically rich characters and committed suicide in March 1941? What is the term used to describe the type of model, such as Betty Grable or Veronica Lake whose images were mass produced so that they could be informally displayed in men's rooms? American poet Emily Dickinson explored which recurring theme in her writing? Jennifer Aniston divorced which Hollywood actor in 2005, amidst allegations that he had an affair with his Mr. and Mrs. Smith co star?
181 Woody Allen Which Jewish screenwriter is noted for quirky comedies, such as Hannah and her Sisters and Midnight in Paris and is also no torious for marrying his Korean stepdaughter, who was 34 years younger? Woodrow Wilson the 28th U.S. President, was known for creating the League of Nations and signing the Treaty of Versailles after which war? Director Roman Polansk i, who fled the U.S. after pleading guilty to having unlawful sex with a minor, won three Oscars for which WWII drama? Former NBA star Shaquille O'Neal is over ___ feet tall and weighed 325 lbs., making him one of the heaviest players to ever play in the league.
182 LIST OF R EFERENCES Abdel Rahman, R., & Melinger, A. (2007). When bees hamper the production of honey: Lexical interference from associates in speech production. Journal of Experimental Psychology: Learning, Memory, and Cognition, 33, 604 614. doi:10.1037/0278 7393 .33.3.604 Abrams, L., & Farrell, M. T. (2011). Language processing in normal aging. In J. Guendouzi, F. Loncke, & M. J. Williams (Eds.), The handbook of psycholinguistic and cognitive processes: Perspectives in communication disorders (pp. 49 73). New York NY: Psychology Press. Abrams, L., & Rodriguez, E. L. (2005). Syntactic class influences phonological priming of tip of the tongue resolution Psychonomic Bulletin and Review 12, 1018 1023. Abrams, L., Trunk, D. L., & Merrill, L. A. (2007). Why a superma n cannot help a tsunami: Activation of grammatical class influences resolution of young and older of the tongue states. Psychology and Aging, 22, 835 845. doi: 10.1037/0882 7922.214.171.1245 Abrams, L., White, K. K., & Eitel, S. L. (2003). Isolatin g phonological components that increase tip of the tongue resolution. Memory and Cognition, 31 1153 1162. Ackerman, P. L., & Rolfhus, E. L. (1999). The locus of adult intelligence: Knowledge, abilities, and nonability traits. Psychology And Aging 14 (2), 314 330. doi:10.1037/0882 79126.96.36.1994 Balota, D. A., & Ferraro, F. R. (1996). Lexical, sublexical, and implicit memory processes in healthy young and healthy older adults and in individuals with dementia of the Alzheimer type. Neuropsychology 10 82 95. Baayen, R.H., Piepenbrock, R., & Gulikers, L. (1995). The CELEX lexical database [CDROM]. Philadelphia: University of Pennsylvania, Linguistic Data Consortium. Besner, D., Moroz, S., & O'Malley, S. (2011). On the strength of connections between localist mental modules as a source of frequency of occurrence effects. Psychological Science 22(3), 393 398. doi:10.1177/0956797610397957 Bock, J. K., & Griffin, Z. M. (2000). Producing words: How mind meets mouth. In L. R. Wheeldon (Ed.), Aspects of language pr oduction (pp. 7 47). New York: Psychology Press. Bredart, S., & Valentine, T. (1998). Descriptiveness and proper name retrieval. Memory, 6, 199 206.
183 Brdart, S., Valentine, T., Calder, A., & Gassi, L. (1995). An interactive activation model of face naming. The Quarterly Journal of Experimental Psychology A: Human Experimental Psychology 48A(2), 466 486. Brennen, T. (1993). The difficulty with recalling people's names: The plausible phonology hypothesis. In G. Cohen, D. M. Burke, G. Cohen, D. M. Burke (Eds .) Memory for proper names (pp. 409 431). Hillsdale, NJ England: Lawrence Erlbaum Associates, Inc. Brown, A.S. (1991). The tip of the tongue experience: A review and evaluation. Psychological Bulletin, 109 204 223. Bruce, V., & Young, A. (1986). Underst anding face recognition. British Journal of Psychology 77(3), 305 327. Burgess, C., & Conley, P. (1999). Representing proper names and objects in a common semantic space: A computational model. Brain and Cognition 40(1), 67 70. Burke, D., Locantore, J., Austin, A., & Chae, B. (2004). Cherry pit primes Brad Pitt: homophone priming effects on young and older adults' production of proper names Psychological Science, 15, 164 170 Burke, D. M., MacKay, D. G., Worthley, J. S., & Wade, E. (1991). On the tip of the tongue: What causes word finding failures in younger and older adults. Journal of Memory and Language, 30, 542 579. Burke, D. M., & Shafto, M. A. (2008). Language and aging. In F. I. M. Craik & T. A. Salthouse (Eds.), The handbook of aging and cogniti on (3rd ed., pp. 373 443). New York: Psychology Press. Burton, A., Bruce, V., & Johnston, R. A. (1990). Understanding face recognition with an interactive activation model. British Journal of Psychology 81(3), 361 380. Carreiras, M., Baquero, S., & Rodrg uez, E. (2008). Syllabic processing in visual word recognition in Alzheimer patients, elderly people, and young adults. Aphasiology 22 1176 1190. Cholin, J., Dell, G. S., & Levelt, W. M. (2011). Planning and articulation in incremental word production: S yllable frequency effects in English. Journal Of Experimental Psychology: Learning, Memory, And Cognition 37 (1), 109 122. doi:10.1037/a0021322 Cohen, G. (1990). Why is it difficult to put names to faces?. British Journal of Psychology 81(3), 287 297. Coh en, G., & Burke, D. N. (1993). Memory for proper names: A review. Memory, 1, 249 263.
184 Cohen, G., & Faulkner, D. (1986). Memory for proper names: Age differences in retrieval. British Journal of Experimental Psychology, 4, 187 197 Conley, P., Burgess, C., & Hage, D. (1999). Large scale databases of proper names. Behavior Research Methods, Instruments & Computers 31(2), 215 219. Cross, E. & Burke, D. (2004). Do alternative names block young and older adults' retrieval of proper names ? Brain and Language, 89 174 181. Cutting, J. C., & Ferreira, V. S. (1999). Semantic and phonological information flow in the production lexicon. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25, 318 344. doi: 10.1037/0278 73188.8.131.528 Damian, M. F., & M artin, R. C. (1999). Semantic and phonological codes interact in single word production. Journal of Experimental Psychology: Learning, Memory, and Cognition, 25, 345 361. doi: 10.1037/0278 73184.108.40.2065 Damian, M. F., Vigliocco, G., & Levelt, W. J. M. (200 1). Effects of semantic context in the naming of pictures and words. Cognition, 81, B77 B86. doi: 10.1016/S0010 0277(01)00135 4 Dell, G. S. (1986). A spreading activation theory of retrieval in sentence production. Psychological Review, 93 283 321. Evrar d, M. (2002). Ageing and lexical access to common and proper names in picture naming. Brain and Language, 81 174 179. Farrell, M. T., & Abrams, L. (2011). Tip of the tongue states reveal age differences in the syllable frequency effect. Journal of Experi mental Psychology: Learning, Memory, and Cognition, 37, 277 285. doi:10.1037/a0021328 Farrell, M. T., Abrams, L., & White, K. K. (2012). The role of priming in lexical access and speech production. In K. Ligouri (Ed.), Psychology of Priming (pp. 205 244). Hauppauge, NY: Nova Science Publishers, Inc. Ferreira, V. S. & Griffin, Z. M. (2003) Phonological influences on lexical (mis) selection. Psychological Science 14, 86 90. Field, A. (2005). Discovering statistics using SPSS. London: Sage. Fogler, K. A., & Ja mes, L. E. (2007). Charlie Brown vs. Snow White: The effects of The Journals of Gerontology: Series B: Psychological Sciences and Social Sciences 62, 201 207. Folstein, M. F., Folstein S. E., & McHugh, P. R. (1975). Mini mental state: A practical method for grading the stat e of patients for the clinician. Journal of Psychiatric Research, 12, 189 198.
185 Garrett, M. F. (1988). Processes in language production. In F. J. Newmeyer (Ed.), The Cambridge survey of linguistics. Vol. 3: Biological and psychological aspects of language (pp. 69 86). Cambridge, MA: Harvard University Press. Gollan, T.H., & Brown, A.S. (2006). From tip of the tongue (TOT) data to theoretical implications in two steps: When more TOTs means better retrieval. Journal of Experimental Psychology: General, 135, 462 483. Hanley, J. (2011). Why are names of people associated with so many phonological retrieval failures?. Psychonomic Bulletin & Review 18(3), 612 617. doi:10.37 58/s13423 011 0082 0 Hasher, L., & Zacks, R. T. (1988). Working memory, comprehension, and aging: A review and a new view. In G. H. Bower (Ed.) The psychology of learning and motivation: Advances in research and theory, Vol. 22 (pp. 193 225). San Diego, CA US: Academic Press. doi:10.1016/S0079 7421(08)60041 9 Hedden, T. & Gabrieli, J.D.E. (2004). Insights into the ageing mind: a view from cognitive neuroscience. Nature Reviews Neuroscience, 15, 87 97. Heine, M., Ober, B., & Shenaut, G. (1999). Naturally occurring and experimentally induced tip of the tongue experiences in three adult age groups. Psychology and Aging 14 445 457. James, L.E. (2006). Specific effects of aging on proper name retrieval: Now you see them, now you don't The Journals of Geront ology: Series B: Psychological Sciences and Social Sciences, 61, 180 183 James, L. E. & Burke, D. M. (2000). Phonological priming effects on word retrieval and tip of the tongue experiences in young and older adults. Journal of Experimental Psychology: Memory, Learning, and Cognition, 26, 1378 1391. doi:10.1037//0278 73220.127.116.118 Jescheniak, J. D., & Levelt, W. J. M. (1994). Word frequency effects in speech production: Retrieval of syntactic information and of phonological form. Journal of Experimental Psychology: Learning, Memory, and Cognition 20 824 843. Johnston, R. A., & Bruce, V. (1990). Lost properties? Retrieval differences between name codes and semantic codes for familiar people. Psychological Research 52(1), 62 67. doi:10.1007/BF00867213 J ones, G.V. (1989). Back to Woodworth: Role of interlopers in the tip of the tongue phenomenon. Memory & Cognition, 17, 69 76.
186 Kemper, S. (2006). Language in adulthood. In E. Bialystok & F. I. M. Craik (Eds.) Lifespan cognition: Mechanisms of change (pp. 22 3 238). New York: Oxford University Press. Kroll, J. F., & Stewart, E. (1994). Category interference in translation and picture naming: Evidence for asymmetric connection between bilingual memory representations. Journal of Memory and Language, 33, 149 174 doi: 10.1006/jmla.1994.1008 Levelt, W. J. M. (1989). Speaking: From intention to articulation. Cambridge, MA: MIT Press. Levelt, W. J. M., Roelofs, A., & Meyer, A. (1999). A theory of lexical access in speech production. Behavioral and Brain Sciences 2 2 1 75. Macizo, P., & Van Petten, C. (2007). Syllable frequency in lexical decision and naming of English words. Reading and Writing 20 295 331. doi:10.1007/s11145 006 9032 z. MacKay, D. G. (1987). The organization of perception and action: A theory for language and other cognitive skills New York: Springer Verlag. MacKay, D. G., & Burke, D. M. (1990). Cognition and aging: A theory of new learning and the use of old connections. In T. M. Hess (Ed.), Aging and cognition: Knowledge organization and utili zation (pp.213 263). Amsterdam: North Holland. MacKay, D. & Burke, D. (1991). Learning and retrieval of names versus other descriptors of people: new theory and data Unpublished paper presented at the International Conference on Memory. Maylor, E. A. (19 95). Effects of aging on the retrieval of common and proper names. In B. J. Vellas, J. L. Albarede, P. J. Garry, B. J. Vellas, J. L. Albarede, P. J. Garry (Eds.) Mood and cognitive disorders (pp. 55 73). New York, NY US: Springer Publishing Co. Maylor, E A. (1997). Proper name retrieval in old age: Converging evidence against disproportionate impairment. Aging, Neuropsychology, and Cognition, 4, 211 226. Maylor, E. A., & Valentine, T. (1992). Linear and nonlinear effects of aging on categorizing and nami ng faces. Psychology and Aging 7(2), 317 323. doi:10.1037/0882 7918.104.22.1687 McWeeny, K. H., Young, A. W., Hay, D. C., & Ellis, A. W. (1987). Putting names to faces. British Journal of Psychology 78(2), 143 149. Meyer, A. S., & Bock, K. (1992). The tip o f the tongue phenomenon: Blocking or partial activation? Memory and Cognition, 20, 715 726.
187 Meyer, A.S. & S chriefers H. (1991). Phonological facilitation in picture word interference experiments: Effects of stimulus onset asynchrony and types of interferi ng stimuli. Journal of Experimental Psychology: Learning, Memory, and Cognition, 17 1146 1160. doi: 10.1037/0278 7322.214.171.1246 Miozzo, M., & Caramazza, A. (1997). Retrieval of lexical syntactic features in tip of the tongue states. Journal Of Experimenta l Psychology: Learning, Memory, And Cognition 23 (6), 1410 1423. doi:10.1037/0278 73126.96.36.1990 Mortensen, L., Meyer, A. S., & Humphreys, G. W. (2006). Age related effects on speech production: A review. Language and Cognitive Processes, 21 238 290. Oldf ield, R., & Wingfield, A. (1965). Response latencies in naming objects. The Quarterly Journal of Experimental Psychology 17 273 281. Pelamatti, G., Pascotto, M., & Semenza, C. (2003). Verbal free recall in high altitude: Proper names vs common names. Co rtex: A Journal Devoted To The Study Of The Nervous System And Behavior 39 (1), 97 103. doi:10.1016/S0010 9452(08)70077 7 Rapp, B., & Goldrick, M. (2000). Discreteness and interactivity in spoken word production. Psychological Review, 107, 460 499. doi: 10 .1037//0033 295X.107.3.460 Sailor, K., Brooks, P. J., Bruening, P. R., Seiger Gardner, L., & Guterman, M. (2008). Exploring the time course of semantic interference and associative priming in the picture word interference task. The Quarterly Journal of Ex perimental Psychology, 62 789 801. doi:10.1080/17470210802254383 Salthouse, T. A. (1996). The processing speed theory of adult age differences in cognition. Psychological Review 103 403 428. Schriefers, H., Meyer, A. S., & Levelt, W. J. (1990). Explorin g the time course of lexical access in language production: Picture word interference studies. Journal of Memory and Language, 29, 86 102. doi: 10.1016/0749 596X(90)90011 N Schweinberger, S. R., Burton, A., & Kelly, S. W. (2001). Priming the access to name s of famous faces. British Journal of Psychology 92(2), 303 317. doi:10.1348/000712601162202 Semenza, C. (2006). Retrieval pathways for common and proper names. Cortex 42, 884 91. Semenza, C. (2009). The neuropsychology of proper names. Mind & Language, 24, 347 369. Shafto, M. & MacKay, D. (2000). The Moses, Mega Moses, and Armstrong Illusions: Integrating language comprehension and semantic memory. Psychological Science, 11, 372 378.
188 Spieler, D. H., & Balota, D. A. (2000). Factors influencing word namin g in younger and older adults. Psychology and Aging 15 225 231. Starreveld, P. A., & La Heij, W. (1995). Semantic interference, orthographic facilitation, and their interaction in naming tasks. Journal of Experimental Psychology: Learning, Memory, and C ognition, 21, 686 698. doi: 10.1037/0278 73188.8.131.526 Starreveld, P. A., & La Heij, W. (1996). Time course analysis of semantic and orthographic context effects in picture naming. Journal of Experimental Psychology: Learning, Memory, and Cognition, 22, 89 6 918. doi:10.1037/0278 73184.108.40.2066 Stemberger, J. P. (1984). Structural errors in normal and agrammatic speech. Cognitive Neuropsychology 1(4), 281 313. doi:10.1080/02643298408252855 Valentine, T., Bredart, S., Lawson, R., & Ward, G. (1991). What's in a name? Access to information from people's names. European Journal Of Cognitive Psychology 3 (1), 147 176. doi:10.1080/09541449108406224 Verhaeghen, P. (2003). Aging and vocabulary score: A meta analysis. Psychology And Aging 18 (2), 332 339. doi:10.1037/ 0882 79220.127.116.112 Vitevitch, M. S., & Sommers, M. S. (2003). The facilitative influence of phonological similarity and neighborhood frequency in speech production in younger and older adults. Memory and Cognition 31 491 504. Vitkovitch, M., Potton, A., Bakogianni, C., & Kinch, L. (2006). Will Julia Roberts harm Nicole Kidman? Semantic priming effects during face naming. The Quarterly Journal of Experimental Psychology 59(6), 1134 1152. White K. K., & Abrams, L. (2002). Does priming specific syllables during tip of the tongue states facilitate word retrieval in older adults? Psychology and Aging, 17, 226 235. doi: 10.1037/0882 7918.104.22.168 White, K. K., Abrams, L., & Frame, E. A. (in press). Semantic category moderates phonological priming of proper name retrieval during tip of the tongue states. Language and Cognitive Processes. Young, A. W., Ellis, A. W., & Flude, B. M. (1988). Accessing stored information about familiar people. Psychological Research 50(2), 111 115. doi:10.1007/BF00309210 Young, A W., McWeeny, K. H., Ellis, A. W., & Hay, D. C. (1986). Naming and categorizing faces and written names. The Quarterly Journal of Experimental Psychology A: Human Experimental Psychology 38A(2), 297 318. Zacks, R. T., & Hasher, L. (1994). Directed ignori ng: Inhibitory regulation of working memory. In D. Dagenbach, T. H. Carr (Eds.) Inhibitory processes in attention, memory, and language (pp. 241 264). San Diego, CA US: Academic Press.
189 BIOGRAPHICAL SKETCH Meagan Farrell was born on March 15 th in For t Myers, Fl. She grew up on Fort Myers Beach with her two siblings, Liam and Caitlin. At Appalachian State University Meagan was starting athlete on the women's soccer team, and received accolades in 2005 as a Division I Academic All American. In May 2006 she graduated summa cum laude with a B.A. in Psychology. Prior to starting graduate school, Meagan worked as a psychometrist in Southwest Florida, conducting testing batteries for patients with a variety of other neuropsychological conditions. After receiv ing a College of Liberal Arts and Sciences Alumni Fellowship, she began her graduate career at UF in the fall of 2007, working closely with her research mentor Lise Abrams. Her research interests are focused on language and memory processes in young and ol der adults, with an emphasis on age related changes to speech production and the retrieval of words from memory. Specific areas of investigation include (i) semantic, lexical, and phonological predictors of word retrieval failures in older age (ii) the te mporal dynamics of lexical access and speech production, and (iii) the neurocognitive basis of language. Her dissertation research was funded by a Jacquelin Goldman Dissertation Fellowship in Developmental Psychology and the William Orr Dingwall Neuroling uistics Dissertation Fellowship. In December 2012 she will be awarded a Ph.D. in psychology with a graduate specialist certificate in Gerontology. Upon completion of her Ph.D, Meagan will continue investigating neuro cognitive change associated with normal aging and age related neuropathology as a post doctoral research fellow at Columbia University College of Physicians and Surgeons.