Kirwan 1 The Effects of Musical Context and Race on Perception and Memory in a Face Recall Task Kirwan, S., Zwaan, T., Hathaway, N., Garas, M., & Scott, L.S. Abstract The environment that our brain perceives, also known as context, has been shown to play a role in influencing cognitive p rocesses such as perception, pain and memory ( Repp, 1982; CDM, e.g., Mead & Ball, 2007 ) The influence of context on memory is referred to as context dependent memory (CDM, e.g., Mead & Ball, 2007). Constant processing of the environment leads to gathering of contextual information which serves as the basis for context dependent memory This current study investigates how context influences our perception and memory by examining how the tonality of music can affect our abi lity to perceive and to recognize faces. The challenge to understand these influences on perception and memory is made easier with the use of e lectroencephalogram (EEG) event related p otentials (ERPs) and behavioral measures involving accuracy and sensitiv ity. The aim of this project is to examine the neural and behavioral correlates of context dependent memory for own and other race faces during a n old/new memory task which manipulated music in order to determine whether encoding or retrieval of faces is influenced by whether the face is from a familiar group and by the context of encoding and retrieval The results revealed an interaction between music context and participant judgment; however, this interaction only occurred when participants made an incorrect judgment (miss or false alarm). This study serves to further our understanding of the brain processes which mediate how context influences our ability to understand and remember the environment.
Kirwan 2 Introduction The other race effect (ORE) is a well studied effect within the field of facial recognition. Research has shown that people have more difficulty in discriminating and remembering faces for races other than their own (Meissner & Brigham, 2001). Consequentl y, there is a higher hit rate for own race faces than for other races faces (Meissner & Brigham, 2001). Although t his effect has been found across cultural and racial groups it is strongest among Caucasians compared to other races (Meissner & Brigham, 200 1). The other race effect is thought to be influenced by many other factors including social categorization, motivated individuation and perceptual expertise (Hugenberg et al., 2010). This theory is known as the categorization individuation model (CIM) (H ugenberg et al., 2010). For example, the theory predicts that adults categorize other race faces and as a result they tend to focus on the facial features shared by people of that race other than their own. This categorizing process is thought to lead to f aces appearing as more homogenous. Motivated individuation is also an important feature of the CIM. Specific cues often motivate individuals to either encode categorical features of a face (i.e. features shared by all faces of that category) or features th at identify individuals (i.e. features that are specific to that face). For example, if a face is perceived to not be relevant to the individual, the individual may encode the face's categorical features more so than the face s individual features. These d ifferent processes in encoding faces may influence the ORE. Lastly, the CIM proposes that experience with facial discrimination plays an important role in the ORE. Because individuals are often raised by people within their own race, they acquire more expe rience discriminating faces of their own race than they do with faces of
Kirwan 3 other races. As a result of this disproportionate experience, individuals become better at discriminating faces of their own race. In order to study face perception, previous experim ents have exa mined specific event related potential (ERP) components related to the p rocess. Components such as the P 100, N170, and N250 have been found to index face processing. The P100 component occurs approximately 100 milliseconds after the stimulus o nset over posterior electrodes and it has been found that faces elicit greater P100 amplitudes than do objects (Hermann et al., 2005). The N170 component occurs approximately 170 milliseconds after the stimulus onset over the right posterior electrodes and faces have been shown to elicit a greater negative amplitude than do non face stimuli (Bentin, Allison, Puce, Perez, & McCarthy, 1996; Rossion, Joyce, Cottrell, & Tarr, 2003). The N170 is thought to be a reliable index of the properties that all faces sha re due to the fact that the race of the face presented has no effect on the amplitude or the latency of the N170 component (Bentin & Deouell, 2000; Eimer, 2000; Caldara, Rossion, Bovet, & Hauert, 2004; Caldara et al., 2003; James, Johnstone, & Hayward, 200 1). In addition to this, it has been shown that the N170 has greater ac tivity for individuals with expertise in identifying the specific stimuli (Scott et al., 2006; Tanaka & Curran, 2001). The N250 components occurs approximately 250 milliseconds after th e stimulus onset and is thought to be an index of face familiarity, because it has been shown to be greater in response to repeated and familiar faces (Schweinberger et al., 2002, 2004; Tanaka et al., 2006). Training participants on model stimuli has bee n shown to influence the ERP components related to face processing (Tanaka et al., 2009). For example, other race training leads to faster peaking in the N170 component post training relative to pre training (Tanaka et al., 2009). Th is suggests faster proc essing of other race faces after repeated exposure. The N250 has also been
Kirwan 4 shown to be greater for faces trained at the individual level relative to faces trained at the categorical level (Tanaka et al., 2009). The N250 was also enhanced for untrained faces within the trained category (Tanaka et al., 2009). The other race effect has also been examined using ERP components related to memory processes (Golby, Gabrieli, Chiao, & Eberhardt, 2001; Stahl, Wiese, & Schweinberger, 2010; Herzmann, Willenbockel, Tanaka, & Curran, 2011) Specifically, two ERP components that index memory processes are the FN400 and the P600. The FN400 has been shown to index familiarity, an aspect of recognition that is fast, yet does not provide precise details related to the memo ry (Rugg & Curran, 2007). The FN400 is measured over the frontal lobe between 300 and 500 milliseconds after the stimulus onset. The P600 is an ERP component found over the parietal lobe of the brain that denotes recolle ction, a slower process that provide s explicit details of the memory (Rugg & Curran, 2007). The P600 is measured between 500 and 800 milliseconds after the stimulus onset. It has been shown that these two ERP components can be used to show the difference between the waveforms of recognized i tems (hits) and correctly rejected new items (correct rejections). It is reported that Asian and Caucasian participants had better memory for own races faces than for other race faces (Herz mann et al. 2011). Specifically, the P600 for Caucasian individuals was found to be influenced by the race of the face presented. For Caucasian participants, o ther race faces elicited a prolonged P600 component over the frontal brain regions, whereas own race faces elicited the typical P600 component (Herzmann et al. 2011 ). This suggests that these participants might show stronger involvement of post retrieval monitoring processes for other race faces (Herzmann et al. 2011).
Kirwan 5 Researchers have also used a Remember Know paradigm to investigate how the other race effect influe nces recollection and familiarity (Horry, Wright, & Tredoux, 2010; Marcon, Susa, & Meissner, 2009; Meissner, Brigham, & Butz, 2005). Own race faces produce a higher "remember" hit rate than do other race faces. This finding suggests that recollection is m ore accurate for own race faces than for other race faces. There were also fewer false alarms found for own race faces. Previous research has found that familiarity is linked to the number of false alarms, therefore the ORE is suggested to influence famili arity processes. The influence of the ORE is only found in false alarm rates and is not seen in hit rates (Diana, Reder, Arndt, & Park, 2006). Memory has also been found to be influenced by changes in context between encoding and testing sessions. This e ffect is known as context dependent memory and can be seen for changes in many different types of settings (CDM, e.g., Mead & Ball, 2007). Memory performance has been shown to be influenced by changes in contexts including: o lfactory changes (Schab, 1990) changes in physical location (Isarida & Isarida, 2004), changes in time of day (Holloway, 1978), changes in mood states (Eich, Macauley, & Ryan), changes in music tempo (Balch & Lewis, 1996) and musical key (Mead & Ball, 2007) During these tasks, memory performance is better for tasks with conditions that are the same during encoding and testing than tasks with conditions that change during encoding and testing. The aim of the current study is to investigate the effects of context, specifically the tonali ty of music, on the memory and perception of own and other race faces. In general, it is predicted that music will influence the memory and perception of faces. This enhanced memory performance will be most effective when the musical context is the same fr om study to test trials. The FN400 P600 and N250 are predicted to have greater amplitudes for
Kirwan 6 same music context trials and own race faces However, the N170 component is predicted to not be influenced by race due to the fact that previous literature has shown race having no effect on the amplitude or the latency of the N170 component (Bentin & Deouell, 2000; Eimer, 2000; Caldara, Rossion, Bovet, & Hauert, 2004; Caldara et al., 2003; James, Johnstone, & Hayward, 2001). Finally, the hit rate, correct reject ion rate and overall percent correct is predicted to be greater for "Same Context" conditions and Caucasian faces due to previous literature on the other race effect and context dependent memory. Materials and Methods Participants 59 undergraduate students from the University of Massachusetts Amherst participated in the study. All subjects were right handed and had normal or corrected to normal vision. All participants used in the study were Caucasian and completed this study for cou rse credit in a Psychology class at the University of Massachusetts Amherst. 8 of the participants we re removed due to EEG artifacts including eye movements, eye blinks, and bad channels. Participants were removed from analysis if they had more than 20% o f trials rejected due to these EEG artifacts. This threshold was decided upon based on previous research (Sawaki & Luck, 2011). In addition to the participants removed due to EEG artifacts, one participant was removed due to a failure to complete the major ity of trials of the experimen t. For behavio ral data analysis, only 55 data files were available for analysis. Stimuli The stimuli used in the experiment consisted of six hundred and seventy nine original photographs of men between the ages of 20 and 35 d isplaying neutral expressions (207 African Americans, 277 Caucasians, and 195 Hispanic faces). Each image was edited in Adobe
Kirwan 7 Photoshop in order to change them to grayscale. A standard face template with identical hairstyle, face contour, and clothing was used to place the internal facial features of the faces so that external cues could be kept constant to promote recognition based on facial features alone (Tanaka & Pierce, 2009). Each image was formatted into a 225 x 311 pixel (1.9 x 2.7 inches) image an d had equal luminance after adjustment using the SHINE toolbox (Wille nbockel, Sadr, et al., 2010) in MATLAB The SHINE toolbox was used to create an average histogram of brightest to darkest pixels for all faces and the faces were then matched for luminanc e using the average histogram The final edited images were presented to participants on a uniform black background on a computer monitor approximately 75 cm from the participant on a visual angle of 12 Â¡ The auditory stimuli used in the experiment were c reated using the "Finale" music composing program. Twenty four novel instrumental piano pieces were created. Twelve of these pieces were created in a major key and the other 12 were created in a minor key. All music clips were continuous and lasted approxi mately seven minutes. One clip of white noise was used for the control conditions. Apparatus and Materials Raw EEG dat a was collected at the University of Massachusetts Amherst using a 128 channel Geodesic Sensor Net connected to a DC coupled 128 channel, high input impedance amplifier (Net Amps 300 TM, Electrical Geodesics Inc., Eugene, OR). The amplified analog voltages (100 Hz lowpass) were collected continuously and digitized at 500 Hz. Each electrode was adjusted until the individual measures of impe dance were under 50 kOhms. ERP data was collected using Net Station 4.4 Acquisition software. This EEG data was processed using Net Station 220.127.116.11 Review and Net Station 18.104.22.168 Tools software and segmented into Event Related
Kirwan 8 Potentials ( ERPs). The mean amplitudes of the ERP components as well as the behavioral data were analyzed using SPSS software. Data organization and pre processing of the behavioral data was performed using RStudio prior to SPSS analysis. Procedure Old/New Recognition Memory Task T he experiment consisted of 17 blocks that included a study phase followed by a corresponding test phase. Each block included equal numbers of own race (Caucasian) and other race (Hispanic and African American) faces. During the study phase participants wer e asked to memorize 16 faces presented serially, 8 of which were Caucasian faces and the other 8 were Hispanic or African American. Each trial within the study phase began with a fixation cross presented on a blank screen for between 1000 to 1200 milliseco nds. Following the fixation cross, the face stimulus was presented for two seconds. After the face stimulus was presented th ere was a one second interval between trials. Following the study phase, there was a 30 second break before the test phase. During t he test phase, the 16 faces from the study phase were presented randomly with 16 new faces not from th e study phase. Each trial within the test phase began with a fixation cross on a blank screen that was presented for between 1000 and 1200 milliseconds. F ollowing the fixation cross, either a novel face or a target face from the study phase was presented. Participants were then asked to report whether they thought the faces presented were "old" or "new" and with how confident they were with their answer ("M aybe", "Likely", "Surely"). The response options appeared on the screen below the face and the order of responses was counter balance. The face stimuli remained on the screen until the participant made a selection from the options. After the participants m ade a selection, they were presented with response feedback
Kirwan 9 displaying either "Wrong" or "Right" depending on whether or not the participant answered correctly. The feedback was displayed for one second before the next fixation cross appeared. Prior to beg inning the actual experiment, participants completed 8 practice trials to ensure understanding of the procedure. Music Context During the study phase an instrumental piano piece in either the major or minor key (between subjects) was played. During the tes t phase instrumental piano music was played again; however, it was either in the same key as the study phase (same context condition) or in a different key from the study phase (changed context condition). In total, there were six blocks of same context co ndition, six blocks of changed context condition and five blocks of the control condition that consisted of white noise There were only five blocks of control condition due to a limited number of face stimuli. During the break between blocks a selection o f bird song was played as a "distraction piece" (adapted from Balch et al.,1996; Mead and Ball, 2007). The purpose of the distraction song was to avoid participants in switched key conditions from simply being "distracted" by the changed key of the second piece compared with participants in reinstated key conditions. ERP and Behavioral Analyses ERP and behavioral measures were used to investigate memory performance and perception during the task The quality of participant's EEG data will be examined to de termine if there is excessive noise in the EEG data caused by excessive eye movements/blinks, muscle movements, or sleepiness. Participants will be removed from analysis if there is deemed to be excessive noise in their data. For participants with good dat a the P100, N170, N250, FN400 and P600 ERP components will be examined across experimental conditions to look for significant
Kirwan 10 differences in perceptual and recollection processes. For behavioral measures, hit rate, miss rate, false alarm rate, correct rej ection rate, overall percent correct, dprime and area under the ROC curve (AUC) were examined across experiment al conditions. RStudio was used to organize behavioral output data files for SPSS analysis. The "tidyverse", "tidyr", "dplyr", "neuropsychology", "ROSE" and "pROC" packages were used in this process (Robin et al., 2011). SPSS software was then used to analyze both ERP and b ehavioral data using multivariate statistical analyses. R esults Behavioral Results For beh avioral data analysis, hit rate, false alarm rate, correct rejection rate miss rate, overall percent correct, d prime and area under the ROC curve were examined. For each of these behavioral measures, a 3 x 2 MANOVA with 3 levels of music context (same co ntext, changed context, control) and 2 levels of race of the face (Own Race, Other Race) was used. False Alarm Rate False alarm rate was calculated by dividing the sum of false alarms by the sum of both correct rejections and false alarms. Analysis rev ealed a significant main effect due to race ( F( 1 54 ) = 4.992, p = 0.030 p 2 = 0.085 ). Other race faces ( Mean = 0.416, SE = 0.013 ) yielded a greater false alarm rate than own race faces ( Mean = 0.382, SE = 0.014) (t(54 ) = 2.266, p = 0.030) Correct Rejec tion Rate Correct rejection rate was calculated by dividing the sum of correct rejections by the sum of both correct rejections and false alarms. Analysis revealed a significant main effect due to race ( F( 1 54 ) = 4.992, p = 0.030 p 2 = 0.085 ). Own race faces (Mean = 0.618, SE = 0.014 ) yield ed a
Kirwan 11 correct rejection rate greater than other race faces (Mean = 0.584, SE = 0.013 ) (t ( 54 ) = 2.266, p = 0.030) (Fig. 1). Area Under the ROC Curve Area under the ROC Curve was calculated using the pROC package in RStudio. Statistical a nalysis in SPSS revealed no main effects; however, there was a marginally significant interaction between race and music condition. Other race faces yielded a greater area under the ROC curve than own race faces than other race fac es for same context condition (t(54) = 1.667, p = 0.096). Hit Rate, Miss Rate, Overall Percent Correct, D Prime Hit rate was calculated by dividing the sum of hits by the sum of hits and misses. Miss rate was calculated by dividing the sum of the misses by the sum of both hits and misses. Overall percent correct was calculated by dividing the sum of hits and correct rejections by the sum of hits, misses, correct rejections and false alarms. D prime was calculated using the neuropsychology package in RStud io by taking the difference of the z score of hits and the z score of false alarms. Statistical a nalysis in SPSS revealed no significant main effects or interactions for hit rate, miss rate, overall percent correct and d prime. Electrophysiological Results Perceptual Components Mean amplitude of the P100, N170 and N250 components were analyzed using two 3 x 2 x 2 x 3 MANOVA s with 3 levels of music context (same conte xt, changed context, control), 2 levels of race of the face ( Own Race, Other Race), 2 levels of judgment (the first MANOVA analyzed hits and correct rejections and the second MANOVA analyzed misses and false alarms), and 3 levels of region (left occipital, mid occipital, and right occipital).
Kirwan 12 P100 Mean Amplitude An alysis found a main effect of region ( F( 2,48 ) = 5.589, p = 0.007 p 2 = 0.189 ) due to the left occipital region (Mean = 2.874, SE = 0.310) having a significantly greater P100 mean amplitude than mid occipital region (Mean = 2.494, SE = 0.285) ( t(48 ) = 2.908 p = 0.017 ) There were also significant two way interactions for the P100 component analysis b etween music context and judgment, race and judgment, and race and region Correct rejections yielded a greater P100 mean amplitude than for hits during the con trol condition ( t(49 ) = 3.495, p = 0.001 ). Other race faces yielded a greater P100 mean amplitude than own race faces when the participants' judgment was a hit ( t(49 ) = 2.169, p = 0.034 ). Correct rejections yielded a grea ter P100 mean amplitude than hits when the participant was presented with an own race face ( t(52) = 2.66, p = 0.010 ). Finally, the left occipital region was found to yield a greater P100 mean amplitude than the mid occipital region for own race faces ( t(48 ) = 3.47, p = 0.003 ). N170 M ean Amplitude Analysis f ound no main effect s for the mean amplitude of the N170 comp onent. However, significant two way interactions were found for music condition and judgment, race and judgment, and race and region. Misses yielded a gr eater N170 mean amp litudes than false alarms during same context test trials ( t(49) = 2.653, p = 0.011 ) (Fig. 2) The changed context condition yielded a greater N170 mean amplitude than the same context condition when the judgment was a false alarm ( t(48) = 2.568, p = 0.040 ) (Fig. 3) Own race faces yielded a greater N170 mean amplitude than other race faces when the participants' judgment was a hit ( t( 49 ) = 3.407, p = 0.001 ). Hits yielded a greater N170 mean amplitude than correct rejections for own race faces ( t(49) = 3.529, p = 0.001 ). Own race faces yielded a greater N170 mean amplitude than other race faces for the right occipital region ( t(49) = 3.509, p = 0.001 ). Own race faces also
Kirwan 13 yielded a greater N170 mean amplitude than other race faces for the mid occipital region (t(49) = 2.400, p = 0.001). Finally, the right occipital region yielded a greater N170 mean amplitude than the mid occipital region for other race faces ( t(48) = 3.269, p = 0.006 ) as well as for own race faces ( t(48) = 4.373, p < 0.001 ). N250 Mean Amplitude Analysis revealed no main effects for the mean amplitude of the N250 component. Howe ver, there were significant two way interactions between music condition and judgment, race and judgment, and race and region The changed context condition yield ed a greater N250 mean amplitude than the same context condition when the participants' judgment was a false alarm ( t(48) = 3.031, p = 0.011 ) (Fig. 3) Other race faces yielded a greater N250 mean amplitude than own race faces when the participants' judgmen t was a hit ( t(49) = 3.741, p < 0.001 ). Hits yielded a greater N250 mean amplitude than did correct rejections for when participants' were presented with an own race face ( t(49) = 2.263, p = 0.028 ). Finally, own race faces yielded a greater N250 mean ampli tude than other race faces in the mid occipital region ( t(49) = 2.243, p = 0.029 ) as well as the right occipital region ( t(49) = 4.0833, p < 0.001 ). Memory Components Mean amplitude of the FN400 and the parietal old/new effect (P600) components were analyzed using two 3 x 2 x 2 x 2 MANOVA with 3 levels of music context (same conte xt, changed context, control), 2 levels of race of the face ( Own race and other race), 2 levels of judgment ( the first MANOVA analyzed hits and correct rejections and the sec ond MANOVA analyzed misses and false alarms ), and 2 levels of region ( the left and right anterior superior regions were used for the FN400 component and the left and right parietal superior regions were used for the P600 component).
Kirwan 14 FN400 Mean Am p litude An alysis found main effect s of rac e and region for the FN400 component as well as significant two way interactions between race and region. Overall, there was a main effect of race ( F(1,49) = 27.860, p < 0.001, p 2 = 0.362 ) due to other race faces ( Mean = 0.413, SE = 0.242 ) yielding a greater FN400 mean amplitude than own race faces ( Mean = 0.038, SE = 0.86 ) (t(49) = 5.255, p < 0.001) There was also a main effect of region (F(1,49) = 29.707, p < 0.001 p 2 = 0.377) due to the left anterior superior region (Mean = 0.573, SE = 0.243) yielding a greater FN400 mean amplitude than the right anterior superior region (Mean = 0.198, SE = 0.251) ( t(49) = 5.460, p < 0.001 ). Other race faces yielded a greater FN400 amplitude than own race faces for the left anterior s uperior region ( t(49) = 5.415, p < 0.001 ) and the right anterior superior region ( t(49) = 3.787, p < 0.001 ). The left anterior superior region yielded a greater FN400 mean amplitude than the right anterior superior region for other race faces ( t(49) = 5.37 8, p < 0.001 ) as well as for own race faces ( t(49) = 5, p < 0.001 ). There was also a two way interaction between music condition and incorrect judgments (misses and false alarms); however, there wer e no significant results after B onferroni corrected pairwi se comparisons ( t(49) = 1.887, p = 0.065 ) Old/New Effect (P600) Mean Amplitude Analysis found a main effect of region ( F(1,49) = 7.048, p = 0.011 p 2 = 0.126 ) due to a significantly greater P600 component for the right posterior superior region ( Mean = 3.789 SE = 0.319 ) than for the left posterior superior region ( Mean = 3.203, SE = 0.272 ) (t(49) = 2. 651, p = 0.011) There was a significant two way interaction between music condition and judgment. Misses were found to yield a greater P600 mean amplitude than false alarms for changed context conditions ( t(49) = 2.951, p = 0.005) (Fig. 4)
Kirwan 15 Discussion Analyses of behavioral data did not reveal any significant main effects or interactions involving music condition. This finding does not support the original h ypothesis that same context music conditions would enhance memory performance observed through hit rate, correct rejection rate and overall percent correct. However, analysis of behavioral data did reveal significant main effects for false alarm rate and c orrect rejection rate due to race. The se findings support previous literature on the Other Race E ffect (Meissner & Brigham, 2001) Participants showed worse memory performance (lower correct rejection rates and higher false alarm rates) when tas ked with re cognizing other race faces. The results of this experiment which tested only C aucasian participants supports t he Other Race E ffect which has been found to be particularly pronounced among Caucasian individuals (Meissner & Brigham, 2001). Although the influence of music context was not seen in the analysis of behavioral data, ERP analysis revealed significant interactions between music context and judgment for multiple ERP components. The eff ected components included N170 N250 and P600. Across all thes e components, music context was found to influence the amplitude of each ERP com ponent when participants made incorrect judgments either misses or false alarms During same context condition s the N170 mean amplitude was greater when the participants' jud gment was a miss than when it was a false alarm ( t(49) = 2.653, p = 0.011 ) The N170 mean amplitude was also greater for the changed context condition than the same context condition when the judgment was a false alarm (t(48) = 2.568, p = 0.040). The N250 mean amplitude was found to be greater for changed context conditions than same context conditions when the participant judgment was a false alarm as well (t(48) = 3.031, p = 0.011 ). Lastly, d uring changed context conditions the
Kirwan 16 P600 mean amplitude was gre ater for when the participants' judgment was a miss than when it was a false alarm (t(49) = 2.951, p = 0.005 ). These findings somewhat support the original hypothesis that music context would have an influence on memory and perceptual ERP components. How ever, i t was unexpected to see significant interactions between music context and participants' incorrect judgments. Perhaps, b rain processes responsible for incorrect judgments predispose participant s to be more susceptible to the influence of context on memory and perception. Changed context conditions may also influence the brain processes responsible for incorrect judgments which would explain the influence on ERP components related to perception and memory during these specific judgments Additional u ne x pected findings of the study included significant two way interactions between race and judgment as well as race and region for the mean amplitude of the N170 component. These findings are inconsistent with previous research which found race to have no influence on the mean amplitude of the N170 component (Bentin & Deouell, 2000; Eimer, 2000; Caldara, Rossion, Bovet, & Hauert, 2004; Caldara et al., 2003; James, Johnstone, & Hayward, 2001) The primary focus of the study was to investigate the role that music context plays in influencing memory and perception. Despite there being no main effects of music condition on any of the perceptual and memory ERP components analysis revealed an unexpected interaction between music condition and incorrect judgment s that effected multip le ERP components related to perception and memory T his finding supports the idea that music context does have an influence on the perception and memory of faces but only during specific situations. Further
Kirwan 17 research should be conducte d to investigate the relationship between brain processes responsible for making incorrect judgments and for integrating context ual information Figure 1. Correct Rejection Rate greater for Own Race faces than for Other Race faces Figure 2 N170 Component greater for Misses than False Alarms during Same Context Condition 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Correct Rejection Rate Race Other Race Average Own Race Average
Kirwan 18 Figure 3 N170 & N250 Components greater for Changed Context Conditions than Same Context Conditions when participant judgment is a False Alarm Figure 4 P600 component greate r for Misses than False Alarms during the Changed Context Condition
Kirwan 19 R eferences 1.Balch & Lewis. (1996). Music Dependent Memory: The Roles of Tempo Change and Mood Mediation. Journal of Experimental Psychology: Learning, Memory, and Cognition vol. 22 (6) pp. 1354 1363 2.Balch, W., Myers, D., & Papotto, C. (1999). Dimensions of Mood in Mood Dependent Memory. Journal of Experimental Psychology: Learning, Memory and Cognition, 25 (1), 70 83. 3.Curran, T. & Hancock, J. (2007). The FN400 indexes f amiliarity based recognition of faces. NeuroImage, 36, 464 471. 4 Eich, E., Maculey, D., & Ryan, L. (1994). Mood dependent memory for events of the personal past. Journal of Experimental Psychology: General, 123, 201 215. 5 Golby, A. J., Gabrieli, J. D. E., Chiao, J. Y., & Eberhardt, J. L. (2001). Differential responses in the fusiform region to same race and other race faces. Nature Neuroscience, 4 845 850. 6 Herzmann, G., Willenbockel, V., Tanaka, J., & Curran, T. (in press) The neural processes u nderlying memory encoding and recognition of own race and other race faces. Neuropsychologia 7 Holloway, F. A. (1978). State dependent retrieval based on time of day. In B. T. Ho, D. W. Richards III, & D. L. Chute (Eds.), Drug discrimination and state d ependent learning 319 343 8 Hugenberg, K., Young, S. G., Bernstein, M. J., & Sacco, D. F. (2010). The categorization individuation model: An integrative account of the other race recognition deficit Psychological Review 117 (4), 1168 1187. doi:10.1037/a 0020463 9 Isarida, T., & Isarida, T. K. (2004). Effects of environmental context manipulated by the combination of place and task on free recall. Memory, 12, 376 384. 10 Mead, K. M. L., & Ball, L. J. (2007). Music tonality and context dependent recall: The influence of key change and mood mediation. European journal of Cognitive Psychology, 19, 59 79 11 Meissner, C. A., & Brigham, J. C. (2001). Thirty years of investigating the own ra ce bias in memory for faces: A meta analytic review. Psychology, Public Policy, and Law, 7 (1), 3 35. 12 Repp, B. H. (1982). Phonetic trading relations and context effects: New experimental evidence for a speech mode of perception. Psychological Bulletin ,92 (1), 81 110. doi:10.1037/0033 2909.92.1.81 13. Robin, X., Turck, N., Hainard, A., Tiberti, N., Lisacek, F., Sanchez, J., and MÂŸller, M. (2011). "pROC: an open source package for R and S+ to analyze and compare ROC curves". BMC Bioinformatics, 12, p. 77 doi: 10.1186/1471 2105 12 77
Kirwan 20 14 Rugg, M. D. & Curran, T. (2007). Event related potentials and recognition memory. Trends in Cognitive Science, 11, 251 257. 15 Schab, F. R. (1990). Odors and the remembrance of things past. Journal of Experimental Psychology: Learning, Memory, and Cognition, 16, 648 655. 16 Sawaki R, Luck SJ. Active suppression of distractors that match the contents of visual working memory. Vis cogn. 2011;19(7):956 972. 17 Stahl, J., Wiese, H., & Schwein berger, S. R. (2010). Learning task affects ERP correlates of the own race bias, but not recognition memory performance. Neuropsychologia, 48 2027 2040. 18 Tanaka, J. W., & Pierce, L. J. (2009). The neural plasticity of other race face recognition. Cogn itive, Affective, and Behavioral Neuroscience, 9 (1), 122 131. doi:10.3758/CABN.9.1.122 19 Willenbockel, V., Sadr, S., Fiset, D., Horne, G. O., Gosselin, F., & Tanaka, J. W. (2010). Controlling low level image properties: The SHINE toolbox. Behavior Rese arch Methods, 42 671 684