<%BANNER%>

Studying the Effects of Social Recommendation Cues on Cognitive Effort Expended in Processing Informed Consent Informati...

Permanent Link: http://ufdc.ufl.edu/UFE0045183/00001

Material Information

Title: Studying the Effects of Social Recommendation Cues on Cognitive Effort Expended in Processing Informed Consent Information for Medical Research
Physical Description: 1 online resource (109 p.)
Language: english
Creator: Takata, Yukari
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: cognition -- consent -- credibility -- elaboration -- experiment -- homophily -- informed -- likelihood -- logistic -- model -- need -- regression -- source
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The purpose of this study was to examine how social recommendation cues influenced cognitive processing of informed consent information for medical research. Contrary to the normative view that social cues promote peripheral processing, results from this study demonstrated that social recommendation cues can significantly increase cognitive effort expended in processing information.  What is more,the effect of social recommendation cues was one mediated by increased motivation rather than a change in perceived ability. This main effect, of social recommendation cues on motivation to process information and cognitive effort, was independent of both need for cognition (NFC) and the passage highlights, which were predicted to enhance readability of the form. Primary study hypotheses were tested using a 2-way factorial analysis of variance (ANOVA).The independent variables were study conditions (standard consent form with no cues, consent form with social recommendation cues, and consent form with highlights only) and need for cognition (low and high) and the dependent variable was cognitive effort expended in processing consent information. The analysis revealed a significant main effect for study conditions, F(2, 107) = 3.23, p = 0.04.  A significant main effect for NFC was also revealed, F(1,107) = 4.02, p = 0.05, and there was no interaction between study conditions and NFC, F(2, 107) = 0.22, p =0.81. Tukey’s HSD post-hoc comparison of the three groups indicated that those in the social recommendation cue condition (M= 4.53, SD = 3.52) expended greater cognitive effort than the highlights-only conditions (M = 2.35, SD = 4.13), p =. 05. Additionally, those with high NFC (M = 4.33, SD = 3.38) reported greater effort expended in processing consent information than those with low NFC (M= 2.96, SD = 4.52).
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yukari Takata.
Thesis: Thesis (Ph.D.)--University of Florida, 2013.
Local: Adviser: Treise, Deborah M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045183:00001

Permanent Link: http://ufdc.ufl.edu/UFE0045183/00001

Material Information

Title: Studying the Effects of Social Recommendation Cues on Cognitive Effort Expended in Processing Informed Consent Information for Medical Research
Physical Description: 1 online resource (109 p.)
Language: english
Creator: Takata, Yukari
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2013

Subjects

Subjects / Keywords: cognition -- consent -- credibility -- elaboration -- experiment -- homophily -- informed -- likelihood -- logistic -- model -- need -- regression -- source
Journalism and Communications -- Dissertations, Academic -- UF
Genre: Mass Communication thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The purpose of this study was to examine how social recommendation cues influenced cognitive processing of informed consent information for medical research. Contrary to the normative view that social cues promote peripheral processing, results from this study demonstrated that social recommendation cues can significantly increase cognitive effort expended in processing information.  What is more,the effect of social recommendation cues was one mediated by increased motivation rather than a change in perceived ability. This main effect, of social recommendation cues on motivation to process information and cognitive effort, was independent of both need for cognition (NFC) and the passage highlights, which were predicted to enhance readability of the form. Primary study hypotheses were tested using a 2-way factorial analysis of variance (ANOVA).The independent variables were study conditions (standard consent form with no cues, consent form with social recommendation cues, and consent form with highlights only) and need for cognition (low and high) and the dependent variable was cognitive effort expended in processing consent information. The analysis revealed a significant main effect for study conditions, F(2, 107) = 3.23, p = 0.04.  A significant main effect for NFC was also revealed, F(1,107) = 4.02, p = 0.05, and there was no interaction between study conditions and NFC, F(2, 107) = 0.22, p =0.81. Tukey’s HSD post-hoc comparison of the three groups indicated that those in the social recommendation cue condition (M= 4.53, SD = 3.52) expended greater cognitive effort than the highlights-only conditions (M = 2.35, SD = 4.13), p =. 05. Additionally, those with high NFC (M = 4.33, SD = 3.38) reported greater effort expended in processing consent information than those with low NFC (M= 2.96, SD = 4.52).
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Yukari Takata.
Thesis: Thesis (Ph.D.)--University of Florida, 2013.
Local: Adviser: Treise, Deborah M.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2015-05-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2013
System ID: UFE0045183:00001


This item has the following downloads:


Full Text

PAGE 1

1 STUDYING THE EFFECTS OF SOCIAL RECOMMENDATION CUES ON COGNITIVE EFFORT EXPENDED IN PROCESSING INFORMED CONSENT INFORMATION FOR MEDICAL RESEARCH By YUKARI TAKATA SCHNEIDER 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 2013

PAGE 2

2 2013 Yukari Takata Schneider

PAGE 3

3 To Logan

PAGE 4

4 ACKNOWLEDGMENTS I thank my parents for opening my eyes to the common bonds people share, my sister, brothers, and husband who continue to explore this earth with me, and friends from Papua New Guinean to the United States who enveloped us with love. I thank Dr. Treise who patien tly directed my ever wandering ways and pushed me to dig ever further to discover the truth, Dr. Weigold who challenged me to continually improve my study design and critically reexamine each step in the research process, Dr. Lewis who fostered for me an o pen atmosphere of learning and exploration, Dr. Oliverio who helped me find my wings in the online world, Mr. Kolb for kindly taking me under his wings and opening doors that were too large for me to open on my own. I also thank the wonderful people with t he UF CTSI biorepository who believed in my project and made my study possible Rosie, Sher, and Melissa who were my ultimate champions Ashley, Akiko, R yu, Matt, and Vivi, for keeping me going toward the finish line, Adam and Marcia for symbolizing all th e fun to come, a nd thank you, Lord, for this beautiful and mysterious world.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ 4 LIST OF TABLES ................................ ................................ ................................ ........... 8 LIST OF FIGURES ................................ ................................ ................................ ........ 9 LIST OF ABBREVIATIONS ................................ ................................ .......................... 10 ABSTRACT ................................ ................................ ................................ .................. 11 C H A P T E R 1 INTRODUCTION ................................ ................................ ................................ ... 13 Research Involving Human Participants ................................ ................................ 13 The Informed Co nsent Process ................................ ................................ ....... 14 Information ................................ ................................ ................................ 14 Comprehension ................................ ................................ ........................ 15 Voluntariness ................................ ................................ ............................ 15 Challenges in Promoting Informed Consen t ................................ .................... 15 Perspectives from Persuasion Research ................................ ............................... 18 2 RELEVANT LITERATURE ................................ ................................ .................... 22 Social Recommendation Cues ................................ ................................ ............... 22 Cognitive Effort ................................ ................................ ................................ ...... 24 Influence of Social Cues on Cognitive Effort ................................ .......................... 26 Normative and Informational Social Influence ................................ ........................ 27 Theoretical Framework ................................ ................................ .......................... 30 Persuasion Route Implications ................................ ................................ ........ 31 Motivation and Ability ................................ ................................ ...................... 31 Factors That Influence Motivation ................................ ................................ ... 32 Personal relevance ................................ ................................ ................... 33 Need for cognition ................................ ................................ ..................... 35 Factors That Influence Ability ................................ ................................ .......... 36 Limitations and Criticisms of ELM ................................ ................................ ... 37 Hypotheses and Research Questions ................................ ................................ .... 40 3 METHODS ................................ ................................ ................................ ............ 44 Participants and Location ................................ ................................ ...................... 44 Recruitment Procedures ................................ ................................ ........................ 46 Study Procedures ................................ ................................ ................................ .. 48 Study Conditions ................................ ................................ ................................ ... 50

PAGE 6

6 Standard Condition with No Cues ................................ ................................ ... 51 Social Recommendation Cue Condition ................................ .......................... 51 Highlights Only Condition ................................ ................................ ................ 52 Measures ................................ ................................ ................................ ............... 53 Independent Variable ................................ ................................ ...................... 53 Need for cognition ................................ ................................ ..................... 53 Sociodemographic information ................................ ................................ .. 54 Manipulation Checks ................................ ................................ ....................... 54 Topic importance ................................ ................................ ...................... 54 Motivation to process information ................................ .............................. 54 Ability to process information ................................ ................................ .... 55 Dependent Variables ................................ ................................ ....................... 55 Cognitive effort ................................ ................................ .......................... 55 Attitude toward overall experience ................................ ............................ 56 Attitude toward presentation format ................................ .......................... 56 Pretest ................................ ................................ ................................ ................... 56 Statistical Analysis ................................ ................................ ................................ 58 4 RESULTS ................................ ................................ ................................ .............. 61 Study Results ................................ ................................ ................................ ........ 61 Scale Reliability Analysis ................................ ................................ ................. 61 Need for cognition ................................ ................................ ..................... 61 Cognitive effort ................................ ................................ .......................... 62 Attitude toward overall experience ................................ ............................ 62 Attitude toward presentation format ................................ .......................... 63 Topic importance ................................ ................................ ...................... 63 Motivation to process information ................................ .............................. 63 Ability to process information ................................ ................................ .... 64 Manipulation Checks ................................ ................................ ............................. 64 Hypothesis Testing ................................ ................................ ................................ 67 Exploration of Research Questions ................................ ................................ ....... 69 5 DISCUSSION ................................ ................................ ................................ ........ 82 Conclusion ................................ ................................ ................................ ............. 83 Implications for Mass Communication Research ................................ ................... 8 4 Theoretical Implications ................................ ................................ ......................... 85 Implications for Informed Consent and Medical Communication Research ............ 86 Limitations and Future Work ................................ ................................ .................. 87 APPENDIX A RESEARCHER SCRIPT AND INSTRUCTIONS FOR PARTICIPANT RECRUITMENT AND ENROLLMENT ................................ ................................ ... 90

PAGE 7

7 B BIOREPOSITORY INFORMED CONSENT FORM WITH SOCIAL RECOMMENDATION CUES ................................ ................................ ................. 92 C STUDY INSTRUMENTS ................................ ................................ ........................ 96 LIST OF REFERENCES ................................ ................................ .............................. 99 BIOGRAPHICAL SKETCH ................................ ................................ ......................... 109

PAGE 8

8 LIST OF TABLES Table page 4 1 Participant cha racteristics ................................ ................................ .................. 71 4 2 Between subject effects for group differences in motivation across study conditions by need for cognition ................................ ................................ ........ 72 4 3 Description statistics of motivation for study conditions by need for cognition ... 73 4 4 Between subject effects for group differences in perceived ability across study conditions by need for cognition ................................ ............................... 74 4 5 Description statistics of perceived ability for study by need for cognition ........... 75 4 6 Between subject effects for gr oup differences in topic importance across study conditions by need for cognition ................................ ............................... 76 4 7 Description statistics of topic importance for study conditions by need for cognition ................................ ................................ ................................ ............ 77 4 8 Between subject effects for group differences in cognitive effort across study conditions by need for cognition ................................ ................................ ........ 78 4 9 Description statistics of cognitive effort for study conditions by need for cognition ................................ ................................ ................................ ............ 79 4 10 Between subject effects for group differences in overall experience across study conditions by need for cognition ................................ ............................... 80 4 11 Between subject effects for group differences in attitude toward presentation format across study conditions by need for cognition ................................ ........ 81

PAGE 9

9 LIST OF FIGURES Figure p age 2 1 Examples of social recommendation cues online ................................ ............. 42 2 2 The elaboration likelihood model of persuasion adapted from Petty & Wegener (1999) in Perloff (2010). ................................ ................................ ..... 43 3 1 Diigo tool bar for iPad ................................ ................................ ........................ 59 3 2 Study conditions ................................ ................................ ............................... 60

PAGE 10

10 LIST OF ABBREVIATIONS ELM Elaboration likelihood model IRB Institutional Review Board NFC Need for c ognition NIH National Institutes of Health

PAGE 11

11 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 STUDYING THE EFFECTS OF SOCIAL RECOMMENDATION CUES ON COGNITI VE EFFORT EXPENDED IN PROCESSIN G INFORMED CONSENT INFORMATION FOR MEDICAL RESEARCH By Yukari Takata Schneider May 2013 Chair: Debbie Tr eise Major: Mass Communication The purpose of this study was to examine how social recommendation cues influenced cognitive processing of informed consent information for medical research. Contrary to the normative view that social cues promote peripheral processing, results from this study demonstrated that social recommendation cues can significantly increase cognitive effort expended in processing information. What is more, the effect of social recommendation cues was one mediated by increased motivati on rather than a change in perceived ability. This main effect, of social recommendation cues on motivation to process information and cognitive effort, was independent of both need for cognition (NFC) and the passage highlights, which were predicted to en hance readability of the form. Primary study hypotheses were tested using a 2 way factorial analysis of variance (ANOVA). The independent variables were study conditions (standard consent form with no cues, consent form with social recommendation cues, and consent form with highlights only) and need for cognition (low and high) and the dependent variable was cognitive effort expended in processing consent information. The analysis revealed

PAGE 12

12 a significant main effect for study conditions, F (2, 107) = 3.23, p = 0.04. A significant main effect for NFC was also revealed, F (1, 107) = 4.02, p = 0.05, and there was no interaction between study conditions and NFC, F (2, 107) = 0.22, p post hoc comparison of the three groups indicated that those in the social recommendation cue condition ( M = 4.53, SD = 3.52) expended greater cognitive effort than the highlights only conditions ( M = 2.35, SD = 4.13), p =. 05. Additionally, those with high NFC ( M = 4.33, SD = 3.38) reported greater effort expended in processing consent information than those with low NFC ( M = 2.96, SD = 4.52).

PAGE 13

13 CHAPTER 1 INTRODUCTION In as early as 1900, Prussia began the practice of securing informed consent from any persons participating in medical research (Vollman & Winau, 1996). This practice was initiated after a physician injected an experimental syphilis vaccine into his patients who were being treated for unrelated conditions. The vaccine failed and infected his unwitting patients, altering the quality of their lives forever Subsequently, informed consent has been one of the central tenants of all modern discussions about ethical research practices. Regulatory highlights concerning the protection of human subjects in medical research include the Nuremberg Code in 1947, which served as a standard for building criminal cases against physicians who participated in the Nazi human experiments, and the Belmont Report in 1974, which was developed in reaction to the Tuskegee Syphilis Studies and also led to the establishment of Insti tutional Review Boards in the United States (Nelson Marten & Rich, 1999). E ach of these ethical milestones came about as reactions to failure in previous systems to elicit voluntary and informed c onsents from human participants and t he issue of promoting i nformed consent s for participants also remains a challenge today (Beskow, Friedman, Hardy, Lin, & Weinfurt, 2010; Degner & Sloan, 1992; Flory & Emanuel, 2004; Paasche Orlow, Taylor & Brancati, 2003). Research Involving Human P articipants According to the Belmont Report, which continues to guide the policies of the National Institutes of Health (NIH) Office of Human Subject and Research (OHSR), research involving human participants is bound to three ethical principles: (1) respect for persons, (2) beneficen ce, and (3) justice (National Commision,1979). The first principle,

PAGE 14

14 respect for persons, requires researchers to recognize each individual as autonomous agents, whose participation is solely determined by their voluntary and correctly informed decision to do so. The second principle, beneficence, holds researchers to, first, do no harm and to work towards maximizing potential benefits while minimizing potential harm when outcomes are less clear. Finally, the third principle, justice, requires researchers t o carefully weigh and disclose groups of people who are predicted to bear the benefits and burdens of their research, primarily because medical studies typically benefit future patients and not participants. Benefits for researchers could also potentially lead to conflicts of interests. These three tenants of medical research encompass the concept of informed consent. The Informed Consent Process Informed consent in medical research is "the person's voluntary agreement, based upon adequate knowledge and und erstanding, to participate in human subjects research or undergo a medical procedure" (National Institutes of Health, 2011). Essential components of the informed consent process include: (1) information, (2) comprehension, and (3) voluntariness according to NIH. Information Current national guidelines enforced by the NIH OHSR require the following minimum components in an informed consent form (National Commission, 1979; Gottesman & Sandler, 2004): Description of the research procedure The purpose of th e procedure Associated risks and benefits

PAGE 15

15 Alternative solutions Consent procedures must also provide opportunity for participants to ask questions and gain more knowledge related to proposed studies. The specific information that is to be communicated is outlined in the informed consent form which is regulated at local levels by Institutional Review Boards (IRB). The IRB is an administrative body charged with protecting the rights and welfare of research subjects through a formalized process of reviewin g, approving, and monitoring any biomedical and behavioral research involving human subjects (National Institutes of Health, 2011). Comprehension Researchers also must adapt the information they hope to convey into a format that is appropriate for peopl e's capacity to understand. Moreover, researchers are obligated to confirm their participants' comprehension of consent information. However, guidelines note that an evaluation of participants' understanding is appropriate but not mandatory. Voluntarines s Voluntariness refers to the fact that people have the right to equally refuse or join research studies without consequence. They may also leave the studies at any time without consequence (National Commission, 1979). Challenges i n Promoting Informed Con sent According to mainstream discussions of the informed consent process, full disclosure of medical information through traditional consent forms remain a challenge because of the complex medical concepts being communicated coupled with low literacy rates among American adults (Paasche Orlow et al., 2003). In 2007, the average reading literacy among American adults was at the eighth grade level (Ridpath,

PAGE 16

16 Greene, & Wiese). Notably, 75% of adults with long term illnesses fell below this national average (K irsch, Jungeblut, Jenkins, & Kolstat, 2003). So, medical researchers are assigned the difficult task of communicating complex information to populations with significant barriers to comprehension. In a review of the informed consent templates of 114 medica l IRBs, researchers found that most templates were written at an average of the 11th grade level. This average was approximately three grades above the IRB's average self recommended standard of 8th grade (Paasche Orlow et al., 2003). This disparity betwe en reading levels of consent forms and literacy levels of potential study participants has fueled a growing body of research suggesting that participants frequently do not comprehend consent information. For example, one study reported that 70% of partici pants in an oncological clinical trial did not know that the experimental treatment had not been proven to be effective (Joffe, Cook, Cleary, Clark, & Weeks, 2001). Even in a clinical trial that largely enrolled participants with college educations, only 1 2% of participants could name the three trial drugs and only 17% could recall three or more out of 23 possible risks (Fortun, West, Chalkley, Shonde & Hawkey, 2008). In addition to issues of participant literacy hampering consent practices, the sheer vol ume of consent information may be limiting participants from making informed consents. A study published in 2011 found that the length of US and international consent forms for NIH sponsored HIV trials were an average of 22 pages, which exceeded recommenda tions for how much information could be reasonably processed (Kass, Chaisson, Taylor & Lohse, 2011). The authors suggested highlighting key information to enhance the reading process may help, but they also noted that the

PAGE 17

17 process of selecting key passages would present an ethical dilemma because all information presented in consent forms are already distilled down to the most important. Additionally, a study evaluating how different stakeholders value information on a consent form found that the IRB offici als identified most of the form as important (72 % of the sentences). But in a telling contrast, participants thought that less than half of the form was vital (40% of the sentences) (Beskow et al., 2010). Importantly, informed consent forms have also been described by patients as symbolic rituals for protecting medical doctors and institutions rather than vehicles for ensuring patient autonomy (Felt et al., 2009, p. 99). Therefore, not only are medical consent forms lengthy and difficult to understand, bu t a discrepancy exists between what different stakeholders perceive to be the true objective of the form and the way they distinguish between meaningful and necessary information. Finally, beyond the issue of information and comprehension, attitudes of bot h medical researchers and participants toward medical decision making and informed consent processes also contribute to the issue of securing informed consent. In a study published in 2009, Felt et al. interviewed stakeholders of a biorepository study th at collected human tissue for research in an effort to take into account broader factors that influenced patient decisions. Stakeholders for this study included medical researchers and patients. Felt and colleagues found that medical researchers viewed in formed consent primarily as a process to secure legal research rather than an issue of assuring the patients' freedom of choice. Comparatively, researchers who also entertained ethically ideal concepts of patient autonomy appeared to justify perfunctory c onsents by believing that patients were not desiring to receive in

PAGE 18

18 depth information: "Often...it [the information] challenges patients very much intellectually. Why should they be interested in something that doesn't really interest them...10, 15 pages, n o patient wants to read that, definitely not [sic]" (Felt et al., 2009, p. 94). When Felt and colleagues interviewed patients following consent procedures, they found that many patients had low interest in learning from the consent process because they w ere more interested in broader notions of contributing to medical progress than specific medical studies. Some patients also avoided asking extra questions about the study because they were concerned about putting further demands on their medical provide r or that refusing a study could cause conflict and complicate their medical care. Another study examining patient preferences for autonomy found that, although they wanted to be informed, patients preferred medical decisions to be made primarily by their medical providers, especially as the severity of the illness increased (Ende, Kazis, Ash, & Moskowitz, 1983). Such findings demonstrate how securing informed consents go beyond the scope of providing accurate and comprehensible information but also delve into realms of social obligation and influence. Perspectives from P ersuasion R esearch Few studies have examined informed consent processes through the lens of persuasion theories Medical literature generally examines informed consent as a freestanding pr ocess and most studies evaluate informed consent in terms comprehension and acquisition of pertinent study information (Felt, et al., 2009; Flory et al., 2004; Sugarman et al., 1998) However, this view fails to account for the larger narrative of social o bligation, and other more symbolic features of social influence that also feed into communication processes (Perloff, 2010 )

PAGE 19

19 On the other hand, persuasion theories, such as the e laboration likelihood model (ELM), examine attitudinal outcomes from a broade r perspective and take into account factors such as peoples' motivations, abilities and the extent of cognitive effort people engage in issue relevant thinking. Indeed, cognitive effort has been shown to be a reliable measure attitudes change in response to messages and to different features related to those messages (Chaiken & Maheswaran, 1994; Eastin, 2001; Petty & Cacioppo,1981) For example, a significant body of research has demonstrated that people who expend high levels of cognitive effort in processing persuasive messages are more likely to base their attitudes on the merits of an argument, form more stable attitudes and are less influenced by counterarguments compared to those who expend low levels of cognitive effort. On the other hand, people who expend low levels of cognitive effort have been shown to base their attitudes on less issue relevant factors, such as the credibility of message sources or the length of arguments as opposed to its merits (Cacioppo, Petty, Fe instein, & Jarvis, 1996) Seeing as how attitudes associated with high levels of cognitive processing reflect the goals for informed consent process es for medical research this study proposes to use cognitive effort as a proxy for comprehension and will u se ELM to help inform how presentation of consent information influences the extent to which the information is processed Furthermore, the Internet has become a valuable medium for communicating medical and health information ( Fox & Jones, 2009 ) Ad vanc ements in technology have paved ways for large numbers of people to create a nd share content online, a nd as content has become more abundant, people have turned to one another to help filter through meaningful information ( Bernhardt & Felter, 2004; Frost &

PAGE 20

20 Massagli, 2008; Kernisan et al., 2010 ; Sundar & Nass, 2001 ) For example, co ns umer recommendations systems help people determine what they consume online from information to products and services Although there is no standard model, consumer recommenda tions systems ge n er ally take in user feedback to generate various symbols intended to communicate an o bject's level of popularity (Lohmann, Z iegler, & Tetzlaff, 2009). Amazon's customer reviews, for example, allow consumers to rate and review products and sellers. Other recommendation systems bring to light popular information, such as Kindle's popular highlights application, which underlines phrases in digital books that have been highlighted most frequently by other Kindle users and displays the number of users that ha ve highlighted each phrase. Hosts of other recommendation systems are helping people share and sort through the ample content found online, and consumer research has demonstrated that user recommendations and ratings significantly influence b ehavioral intentions and attitudes (Park, Lee, & Han, 2007; Lohmann, Ziegler, & Tetzlaff, 2009 ). Significantly ELM, which has been instrumental in isolating and identifying features of messages that promote or diminish cognitive processing, has generally identified social cues, such as those used in recommendation systems, as features that diminish cognitive processing ( Petty & Cacioppo,1981; Petty & Wegener, 1999; Perloff, 2010; Baran & Davis, 2009 ). ). However as social interactions have become more com monplace online and more applications have come to suppor t and promote these interaction s a new theme has begun to emerge from research examining how people impact o ne another online : the opinions and recommendations of other people offer more than simple strategies for sorting information they also elicit responses that are

PAGE 21

21 more closely tied to age old forces of social influence (Fox & Jones, 2009Hu & Sundar, 2010; Walther, DeAndrea, Kim, & Anthony, 2010; Wang, Walther, Pingree, & Hawkins, 2008) Conseq uently, the impact of these social cues may have greater influence on cognitive effort than what is currently accepted in the literature Therefore, the purpose of this study is not only to examine how presentation formats of informed consent information i nfluence cognitive processing, but more specifically, to examine how social recommendation cues influence cognitive processing of informed consent information for medical research.

PAGE 22

22 CHAPTER 2 RELEVANT LITERATURE Social Recommendation Cues S ocial recomme ndation cues are symbols, such as dots, stars, and numbers that signal how favorably peers recommend products, services, or information online ( Lohmann, Ziegler, & Tetzlaff, 2009; Tanis & Postmes, 2003 ) Social cues are more generally symbols that hint at social interactions and are often used in psychological research to identify how subtle forces of social influence impact attitudes and behaviors ( Petty & Cacioppo,1981; Petty & Wegener, 1999; Perloff, 2010; Tanis & Postmes, 2003 ). For example, in an exper iment examining how peer recommendations influenced participant attitudes toward online products, researchers used dots as social cues to communicate how favorably peers recommended products : a h igh number of dots indicated high favorability and a low numb er of dots indicated low favorability. Participants were also given product images and specifications to help form their judgments; however, researchers found that participant judgments toward products were most significantly influenced by favorability of peer recommendations While o ther factors, such a social trust and product categories, were also important mediators of this interaction this study demonstrate d how social cues can be influential even when others are not physically present but only impli ed or imagined ( De Vries & Pruyn, 2007; Latan,1981) Social cues are thought to exert their influence through a b asic principle of social psychology in which people look to others to resolve uncertainty in their judgments (Festinger, 1954). P eople conti nuously adapt to these cues, including online typographic elements, because adapting to social cues help s people better interpret and interact with their social environments ( Denne t t, 1987; Latan,1981 ).

PAGE 23

23 E xamples of social recommendation cues include Amaz which allow consumers to use stars to rate products and sellers and p opular h ighlights application, which underlines phrases that have been highlighted most frequently by other Kindle users, along with the number of people who highlighted each phrase (Figure 2 1 ). Diigo, a social bookmarking application, also uses social cues to allow users to markup and share any part of a webpage, on any website. Diigo users can highlight passages on a webpage, add sticky notes, and excha nge comments through a special link forwarded to another user These social applications help people form judgments based on the opinions of others. Because all information in consent forms is there for a reason, peer recommendation cues for consent inform ation will be similar to that used in both Kindle and highlighting application s This ensures that the basic content remains intact, while also allowing for the incorporation of the peer recommendations. Therefore for this present research, a s oci al recommendation cue is defined as a set of symbols, such as highlights and number of peer recommendations, that are integrated into the basic format of the informed consent document. According to findings in consumer research, user recommendations signif icantly influence behavioral intentions and attitudes This was best highlighted in an experimental study investigating the influence of online reviews on customer purchase intentions by Park, Lee, and Han (2007) Their research demonstrated that the qual ity and number of reviews have a direct relationship with consumer purchase intentions. Participants in the group exposed to a low number of poor quality reviews demonstrated the least intentions to purchase. In contrast, the group with a high number of hi gh

PAGE 24

24 quality reviews demonstrated the highest intentions to purchase (review quality was pretested in focus groups) Most importantly, the number of reviews, regardless of their quality, most significantly predicted purchase intentions and perceived trustwor thiness of the reviews. Similar studies by both Chevalier and Mayzlin (2006) looking at consumer ratings and book sales, and Zhang and Dellarocas (2006) looking at the movie industry have obtained similar results. T he literature suggests that consumers prefer user generated reviews to seller reviews because they are seen as more consumer centered reviews (Dellarocas, 2003; Bickart & Schindler, 2001). Additionally, Pew Internet's 2009 report on the social life of health information reported that 24% of adults who looked online for health information consulted rankings or reviews for doctors or other providers and that 24% consulted rankings or reviews of hospi tals or other medical facilities (Fox & Jones).Therefore, indirect evidence that social recomme ndation systems and associated social cues influence attitudes is found in many disciplines but less is known about how it influences the cognitive effort expended in processing issue relevant information. Cognitive E ffort People are said to have limited cognitive resources which prevents them from attending to all information in their physical and social environment ( C acioppo et al., 1996 ). Nevertheless, people are continuously allocating their limited cognitive resources to assimilate incoming social in formation in to existing knowledge structures, to form attitudes, and to impart meaning to their worlds (Osgood, Suci, & Tannenbaum, 1957). T his process of assimilating information involves evaluative reactions (like/dislike, helpful/ harmful, etc.) that pe ople generate in response to the experience It is this continual reflection in relation to stimuli

PAGE 25

25 and events relevant thoughts as being the pr Stoltenberg, 1985, p. 227). In communication and persuasion research, cognitive effort which is also called elaboration, d Wegener et al., 1995 ). As mentioned, a ttitudes resulting from high levels of cognitive effort have been characterized as informed, stable, resistant to counterargument, and more pr edictive of behavioral change/intentions compared to judgments resulting from low levels of cognitive effort ( Perloff, 2010; Wegener et al., 1995 ). According to Cacioppo, Petty, and Stoltenberg (1985), high cognitive effort is an indicator of attitude stab ility because resulting attitudes are based on thoughts and associations that are central to the attitude object. Moreover, attitudes resulting from evaluations that involved high cognitive effort are expected to be predictive of behavior because people ar e more likely to be confident in their judgments and willing to act on their formed attitudes (Cacioppo, Petty, & Stoltenberg, 1 985 ). High cognitive effort has also been associated with greater recall of information (Cacioppo, Petty, & Morris, 1993) and c omprehension ( Piolat, Olive, & Kellog, 2004 ). Importa ntly, these attitudinal outcomes precipitated by high cognitive effort reflect fundamental goals of the informed consent process for medical research. For this reason, a study of cognitive effort in the setting of informed consent will provide key insights into participant attitudes and behaviors in response to the informed consent process.

PAGE 26

26 Influence of Social Cues on Cognitive Effort Social cues and the social forces of influence that they embody are n ot customarily viewed as stand alone agents for increasing cognitive effort in persuasion research ( Petty & Cacioppo,1981; Petty & Wegener, 1999; Perloff, 2010; Baran & Davis, 2009 ). Instead, social cues are associated with low cognitive effort and often r eflect what people attend to when motivation and ability to process iss ue relevant information is low For example, a large body of research has demonstrated that when people evaluate persuasive communication that is of low personal relevance, their motiv ation to process information is low, so people will generally support recommendations offered by expert sources, regardless of argument quality (Chaiken& Maheswaran, 1994; Eastin, 2001; Petty & Cacioppo,1981) Such cognitive strategies, allow people to sav e their limited cognitive efforts for when topics are more personally relevant, i n which case people will critically evaluate issue relevant information by expending their reserve of cognitive effort and will base their judgments on argument quality, regar dless of source expertise. These findings demonstrated why social cues have become associated with low cognitive effort. However, a notable feature of these studies is that the social cue, often represented by source expertise, is only considered a proxy for low cognitive effort while argument quality is considered a marker of high cognitive effort. Such an approach juxtaposing social cues against argument quality has become ubiquitous that it is possible that social cues have become associated with l ow cognitive effort by virtue of how they have been operationalized and not necessarily because of the nature of social influence. What is more, no studies have examined the application of s ocial cues as an independent factor meant to increase cognitive pr ocessing. Therefore, little

PAGE 27

27 is known about whether or not social recommendation cues could facilitate high cognitive effort. This is especially true in research areas focused on determining how various features of the communication environment influence th e nature and amount o f issue relevant thinking Normative and Informational S oc ial I nfluence T he dominant view of social influence has been one of conformity and yielding to the judgments of others, but r esearchers seeking to distinguish between different forms of social influence have helped to cast a more positive light upon the impact of social influence. The prevailing view of social influence is best demonstrated through the Asch conformity experiments. In these experiments conducted by Solomon Asch ( 1951, 195 5 ), participants were asked to take part in a simple vision test by selecting a line that matched up with the length of another (participants could select the matching line out of a group of three lines). Each participant was asked to undergo thes e trials 18 times and to select a matching line each time. The task was easy and when participants performed the task alone, they gave the correct responses the vast majority of the time (mistakes were made in only 1% of all trials). However, these experi ments were about conformity, so the power of social influence came to light when participants were surrounded by peers peers who were in fact confederates and had been instructed to give the same but incorrect response in 12 of the 18 trials. The purpose of this manipulation was to see the extent to which participants yielded to the majority of peers Asch discovered that 75% of the participants agreed with their peers and selected the incorrect answer during at least one trial resulting in a 36.8% incorr ect response rate overall The influence of peers has been recapitulated in other fields of research as well ( Cruz, Boster, & Zuckerman, 1999 ; Henningson, Henningson, & Morrill,

PAGE 28

28 2003 ; Gigone & Hastie 1993 ; Isenberg 1986 ; Wittenbaum, Hubbel, & Zuckerman, 1999 ; Janis 1972). argued that two types of influence, normative and informational, were operating in normative social infl uence informational social influence orms of influence are often found together but that it was possible that people conformed to the opinions of others because they accepted it as evidence about reality, regardless of whether they had a sense of social obligations to do so. Essentially, t hey based their argument on a basic tenet of social psychology : people look to others to resolve uncertainty in their judgments (Festinger, 1954) To test their proposition that there were two forms of influence, Deutsch and 1) experiment with a few modifications T o help distinguish between normative and information influence, they included manipulations for uncertainty and certainty concerning line judgments. Specifically participants in the certainty condition were given paper to record their judgments before they were asked to make their judgments public, while participants in the uncertainty condition were not given paper to record their judgments. Significantly, those who had committed to judgments on paper made fewer mistakes than those who did not make commitments informational social influence, and not just conformity and social pressure, was

PAGE 29

29 ertain participants were about their judgments, the more influenced they were by i nformational social influence. Deutsch and Gerard (1954) also demonstrated that normative social influence was at play when they found that those in a face to face condition were face to face) made significantly more mistakes than those in an anonymous condition (participants were separated by partitions so that their only form of interaction with other participants (i.e. confederates) was through lights o n panels that indicated their line judgments). Nonetheless the influence of social peers was significant regardless of anonymity given the high error rates in both groups. Although c ertainty surrounding private judgments (writing the answer down in adva nce) predicted that the subjects would not yield as frequently to social influence the ability to write down the answer in advance did not complete ly eliminate the effect of the real and perceive peer influence Therefore, Deutsch and Gerard (1954) reveal ed that informational social influence is a powerful force that helps people ascertain what is real in their social environments and that normative influence, or the desire to conform to the expectations of others, was not the only force that lead to confo Reflecting on informational social influence, Deutsch and Gerard (1954) same objective situation are discrepant, each will tend to re examine his own view and that of the others t o see if they can be reconciled ( p.635 ). Recalling how these self evaluations are at the heart of what gene rate cognitive effort it is probable that social recommendation cues can motivate high cognitive effort through informational social influence.

PAGE 30

30 Theoretical Framework The e laboration likelihood model (ELM) has been instrumental in revealing how different features of messages influence cognitive processing ( Petty & Cacioppo,1981; Petty & Wegener, 1999 ; Perloff, 2010; Baran & Davis, 2009 ). This is because ELM is optimize their lives but that they have limited cognitive resources to attend to all information in t heir social environment. Therefore, ELM proposes that people use different cognitive selection strategies to cope with their limited cognitive resources. As a result, research based on ELM is aimed at examining the differences between what people perceive to be rational and that which is objectively rational ( Cacioppo, Petty, & Stoltenberg, 1985 ). The e laboration likelihood model (ELM) is a typological, explanatory, and predictive theory of persuasion. It is a dual processing model which holds that people employ two information processing strategies depending on how much motivation and ability they have to process information. The two routes are called the central and peripheral routes to persuasion. Central processing refers to the route employed by those who critically examine and evaluate the merits of issue relevant information to influence their attitude (e.g. merits of an argument). Conversely, peripheral processing is employed by those who do not attend to the issue relevant communication but rathe r non issue relevant features of the communication (e.g. length of argument, source expertise, and attractiveness of the speaker) (Axsom & Chaiken, 1979; Chaiken, 1979; Kalick, 1988). Key tenants of ELM are displayed in Figure 2 2

PAGE 31

31 Persuasion Route Implica tions The routes people employ to form attitudes are also significant because the routes are predicative of attitude stability: those who employed the central route tend to maintain more consistent attitudes and are less influenced by counterargument; whe reas those who employed the peripheral route tend to have less consistent attitudes and are more susceptible to counterargument (Cacioppo, Petty, Feinstein & Jarvis, 1996; Petty et al., 1993). The impact of processing routes on attitudes and behaviors als o has been demonstrated in medical literature (Steginga & Occhipinti, 2004; Trumbo, 1999). In one study, community members in a suspected cancer cluster were surveyed about their perceived risk for cancer, their understanding of the situation, and how the y were seeking more information (Trumbo, 1999). Those who employed the central process had perceptions of higher risks compared with those who used the peripheral process. In other words, the processing routes that people use also influence the level of pe rceived risk and information sought, in the face of high risk Therefore, the most noteworthy link between the e laboration likelihood model and the issue of securing informed consent is the match between predicted and desired outcomes: central route proce ssing conceptually leads to informed consents, and peripheral route processing leads to uninformed consents. Motivation and Ability Motivation and ability are key antecedents in ELM that influence processing routes that people employ. Motivation refers to intrinsic characteristics of people and their drive to achieve a prescribed goal (Ryan & Deci, 2000). Ability refers to actual mental faculties to process information and environmental factors that influence people's capacity to attend to issue relevant i nformation. Generally, the more amenable

PAGE 32

32 the communication environment and messages are to promoting motivation and ability, the more likely people are to take the central processing route, and vice versa (Cacioppo et al., 1996; Petty et al., 1993). For ex ample, an experimental study by Petty and colleagues (1981) demonstrates how motivational factors influence attitudes and decisions. In their study, college students were asked to assess a series of arguments that proposed a mandatory and comprehensive ex am for those graduating from their program. The purpose of this task was to investigate the interacting effects of argument quality, source expertise, and personal relevance on decision making. The p ersonal relevanc e of the issue was manipulated by proposi ng to implement the exam either within the students' academic career or for their distant successors. Additionally, the source's level of expertise was manipulated in the following manner: proponents of the policy were either a class of high school student s or an academic commission. Results of their study showed that when the policy was proposed at a later time (low personal relevance), students expressed more approval for the policy if the message was presented by an expert. But in contrast, they opposed the policy if the message came from high school students, regardless of the quality of the arguments. However, when the proposed policy was to be implemented during their tenure (high personal relevance), students based their decisions on the quality of t he arguments, regardless of the source expertise. This study highlights how personal relevance affects motivation to process information and subsequently formed attitudes. Factors That Influence M otivation The first antecedent in ELM that influences the pe rsuasion route is motivation (see Figure 1). Factors that influence people's motivation to process persuasive communication include: personal relevance, object importance, and n eed for cognition

PAGE 33

33 (NFC). Personal relevance is "the extent to which people bel ieve that a topic or attitude object holds significant consequences for some aspect of their lives" (Cohen, Stotland & Wolf, 1955; Wegener et al., 1995). Object or topic importance, on the other hand, is positively associated with personal relevance but re fers to the amount of personal importance one ascribes to an attitude object rather than the attitude itself (Wegener et al., 1995). N eed for cognition (NFC) is a person's dispositional tendency "to engage in and enjoy effortful analytic activity" (measura ble using an 18 item n eed for cognition Scale ) ( Cacioppo, Petty & Morris, 1983; Cacioppo et al., 1996). People with low NFC tend to rely on peripheral features of persuasive communications, and those with high NFC tend to rely on central features and so ar e less swayed by the peripheral features (Cacioppo et al., 1983; 1996). Personal r elevance Personal relevance of messages is especially important for enhancing motivation to process information. Personal relevance is generally based on the consequence s or effect s an issue could have on an individual. M essage sources and the way messages are framed can also influence how personal ly relevant a topic is perceived to be ( Liberman, & Chaiken, 1996 ). For example, user generated information online has been report ed to increase personal relevance because user information is thought to be audience centered and trustworthy (Bernhardt & Felter, 2004; Frost & Massagli, 2008; Kernisan et al., 2010). Additionally, h omophily is another construct related to personal rele vance Homophily is the tendency for people to unconsciously and consciously show favor toward people that they perceive to be similar to themselves It has been shown to influence how people respond to online health information, including health tips

PAGE 34

34 (McC rosky, Richmond & Daly, 1975 ; Suzuki & Calzo, 2004; Wang et al., 2008; Walther, DeAndrea, Kim & Anthony, 2010). For example a n experimental study conducted by Wang et al. (2008) revealed that tips for coping with chemotherapy posted by a layperson in a di scussion board had more effect on health intentions compared with the same information posted as a tip on a professional website. Originally, source credibility was hypothesized to be the driving influential force for institutional health websites and homo phily for health related discussion boards However, homophily was the most significant driving force for both credibility and likelihood to act on advice for both forums (th e website and discussion group) T he authors also conclude d that chemotherapy tips posted by lay users affect attitudes more strongly than those posted by experts. Further support for the influence of homophily online was provided by a psychosocial experiment exploring the effects of user comments and expert information presented simu ltaneously on a webpage. In a direct comparison of the perception of official and lay messages, Walther and colleagues (2010) explored the interaction between user comments on YouTube and anti marijuana public service announcements (PSA). Overall, the auth ors discovered that valence of user comments (negative or positive) had a stronger effect on attitudes toward the PSA than the actual PSA videos themselves. Most strikingly, the level of homophily expressed in response to user comments was the only signifi cant factor that predicted attitudes toward the health risks of marijuana. Therefore, findings in health communication literature also support the idea that homophily is important in establishing personal relevance for messages encountered online

PAGE 35

35 Need f o r c ognition Need for cognition (NFC) is used in persuasion research to identify people's natural tendency to enjoy and engage in thinking (Cacioppo & Petty, 1982; Cohen et al., to think (intelligence) and is especially useful for examining how different people respond to persuasive communication. A significant body of research has characterized those with high need for cognition (NFC) as people who better attend to communication recall more topic relevant points, and as a result, are less susceptible to peripheral cues that are known to unduly influence behavior (Petty et al., 1996). Examples of peripheral cues include one's reliance on other people and their perceived credibili ty (such as celebrities and experts) instead of the validity of an argument for forming attitudes. On the other hand, those with low NFC are characterized as "cognitive misers" and they are more likely to rely on peripheral cues to form decisions rather t han processing topic relevant information (C acioppo et al., 1996, p. 197). This low tendency for informational processing has prompted researchers to explore mechanisms of improving motivation. A variety of methods have been shown to motivate those with l ow NFC to better evaluate topic relevant information. Two commonly used strategies include increasing personal relevance of a given topic by directly altering potential consequences (Axsom, et al., 1987; Petty, et al., 1981) and framing the message accord ing to one's motivational framework ( Evans & Petty, 2003 ) For example, in a n experimental study examining how messages framed to accommodate NFC would influence medical decision making messages about two cancer treatments were framed in terms of its rat es of survival or rates of death The authors predicted that low NFC participants would opt for the treatment option with the lowest death rate during

PAGE 36

36 treatment, even though death rates were higher at the end of the year compared to the other treatment opt ion. Their hypothesis was supported and the effect of framing was only noted in participants with low NFC In contrast, high NFC participants were unaffected by framing, showing that they were less susceptible to attempts to alter their frame of reference (Smith & Levin, 1996). Similar findings were also reported by Williams Piehota Pizarro and Schneider (2003). In their study mammography screening messages that were matched to individual NFC motivated more mammographies six month later than those that were not intentionally matched Therefore, changes in message features appear to especially influence those with low NFC and the level of cognitive effort they expend. Accordingly, the independent influence of NFC on cognitive pr ocessing is a feature that should be addressed when studying decision making that could be impacted by peripheral processing Factors That Influence Ability The a bility to process available information is another antecedent to determining processing routes. According to ELM, factor s that influence ability include: prior knowledge of issue relevant information, general intelligence, distractions, message clarity, and message comprehension. W hen people perceive information to be too complex for their abilities or situation, they ten d to default to peripheral processing rather than exercising greater elaboration. For example, lengthy issue relevant information has been shown to discourage careful evaluations, resulting in people basing their attitudes and decisions on the length of th e persuasive communication, rather than the merits of arguments contained in the communication (Brock & Green, 2005). Recalling the lengthy and complex nature of consent forms and the lack of opportunity participants are given to process information, ELM p ostulates that

PAGE 37

37 participants would resort to the peripheral route to make their decisions. As predicted, consent behaviors observed by Felt et al. (2009) matched those conjectured by ELM: Patients were asked for consent at a moment when all attention was directed towards the surgical intervention. As a result, instead of assessing the donation request based on the fairly complex written information in the consent form, most con tent of the form into a language that she assumed was suitable for laypeople; that is, the medical student created parallel narratives to the written form. As a consequence, patients hardly read the form; they only browsed through the document because they were requested to sign every page. The patients asked virtually no questions, and, in the end, every patient agreed to donate skin tissue to research. (p. 95) Also, according to the medical literature, chronic illnesses among older adults, such as diabete s and high blood pressure, are associated with reductions in cognitive abilities ; therefore, low cognitive processes may be more prevalent among patient populations (Zelinksi, Crimmins, Reynolds & Seeman, 1998). In this setting, e nhancing con struct may be important for facilitating informed consent. Additionally, Kass, Chaisson, Taylorm, and Lohse (2011 ), in their study of the lengthiness of consent forms, it was suggested that highlighting key information in consent forms could help enhance t he reading process Expanding upon this idea, associating highlights with a social recommendation cue could further improve information processing; therefore, the independent influence of highlights should also be examined along with the influence of soci al recommendation cues Limitations and Criticisms of ELM ELM is one of the most widely used models in persuasion and communication research but also has some unresolved criticisms (Perloff, 2010). The primary criticism of the ELM is that it does not clea rly distinguish the relationship between peripheral and central features of communication processing. For instance, can people use both

PAGE 38

38 routes to persuasion simultaneously? If so, does this not suggest the two routes are in fact one? Founders of the model have addressed this dilemma by explaining that the model and the two routes are based on the two extremes of the elaboration continuum, which inherently implies a relationship between the routes, but that the model is intended to account for the effects o f variable extremes. Essentially, the founders concede that the whole breadth of the elaboration continuum, specifically those that fall in the middle, is not accounted for in the model (Petty et al., 1993). T heir response does not clearly offer an answer to those who argue that "extremes" imply oneness of the underlying concept of elaboration, which in turn, imply that there is but one route to persuasion (Choi & Salmon, 2003). But taking into account that the models and theories are intended to breakdow n larger constructs into more manageable and interlacing concepts for researchers to manage, the issue of whether the two routes to persuasion are one and the same appears to be less important in light of the significant findings consistently generated by the model. ELM is also limited because of its dependence on experimental methods and student populations. Although these approaches lead to higher internal validity, the tradeoff is lower external validity. However, the weight of this criticism may be dam pening by its more recent successful application in other fields of study and associated audiences, such as health campaigns (Williams Pietah, et al., 2003). Also, analysis of variance (ANOVA) had been the standard method of analysis for the model and is i nherently less effective at accounting for covariates and confounding variables, but more recent uses of more rigorous statistical methods of analyses has generally supported the basic constructs of the model (Evans & Petty, 2003; Trumbo, 1999).

PAGE 39

39 Finally, E LM has been criticized for placing significant attention on attitudes and not direct behaviors. However, its harmony with prominent behavioral theories, such as the Social Cognitive Theory (SCT), may suggest that this attitude behavior link may be stronger than what is explicitly stated in ELM related research. According to the Social Cognitive Theory (Bandura, 2001), human behavior is learned as people observe and model the behaviors of other people. And learning is more likely to occur if there is a close identification with the behavior model. This identification construct is akin to the similarity and credibility constructs in ELM. Furthermore, self efficacy, or a person's perceived ability to perform a behavior, is said to further motivate behavioral a doptions, which is especially similar to the motivation and ability constructs in ELM. Given the concordance between the behavior centered SCT and the attitude centered ELM, the influence of attitudes as predictors of behavior could be greater than what i s directly stated in academic literature.

PAGE 40

40 Hypotheses and Research Questions Based on literature reviewed in the previous section, the following hypotheses and research questions are proposed. First, the primary purpose of this study is to examine how soc ial recommendation cues influence cognitive effort expended in processing consent information, so the first hypothesis is proposed, H 1 : Consent forms with social recommendation cues will increase cognitive processing compared to consent forms with no cues Additionally, ELM suggests that the highlights associated with the social recommendation cue could enhance ability to process information independently from the social cue T herefore, the next two hypotheses are proposed in order to isolate the influenc e of the social recommendation cue from highlights and to account for the H 2 : Consent forms with social recommendation cues will increase cognitive processing compared to consent forms with only highlighted passages. H 3 : Consent forms with only highlighted passages will increase cognitive processing compared to consent forms with no cues. Furthermore ELM suggests that need for cognition (NFC) could also have significant influence on cognitive p rocessing. This is because NFC has been applied to diverse contexts and has been shown to independently predict cognitive effort expended in processing issue relevant information and resulting attitudes (Cacioppo & Petty, 1983; Cacioppo et al., 1996; Evans & Petty, 2003; Williams Piehota et al., 2003 ). In order to examine how NFC influences cognitive processing of consent information and how the informed consent process match es up to basic predictions made by ELM, the following hypothesis is proposed,

PAGE 41

41 H 4 : NFC is positively related to cognitive effort that people expend in processing consent information, with those with low NFC exercising less cognitive effort than those with high NFC. What is more those with low NFC may be influenced by social recommend ation cues more greatly than those with high NFC for two reasons: 1) those with high NFC already have a high threshold for cognitive processing so the influence of the social recommendation cue, even if it is positive may be less detect able among those wi th high NFC, and 2) those with low NFC are characterized as people who depend on peripheral cues which include recommendations by credible sources, so those with low NFC may be especially influenced by social recommendation cue s Therefore, the following two hypotheses are proposed H 5 : S ocial recommendation cue s will increase cognitive processing of informed consent information among those with low NFC H 6 : Social recommendation cues will have no impact on cognitive processing among those with high NFC. Finally, few studies, if any, have used social recommendation cues and associated online applications to help communicate informed consent information, so the following researcher questions are asked: RQ1: Do people exposed to recommendations made by oth er people evaluate their overall experience differently than those who are not? RQ 2 : How do participants evaluate the use of social recommendation systems to help communicate informed consent information?

PAGE 42

42 Figure 2 1. Examples of social recommen notes.

PAGE 43

43 Figure 2 2. The elaboration likelihood model of persuasion adapted from Petty & Wegener (1999) in Perloff (2010).

PAGE 44

44 CHAPTER 3 ME THODS The purpose of this study was to examine how a social recommendation cue influences cognitive processing of medical informed consent forms. T he influence of the social recommendation cue on cognitive processing of consent information was tested again st the standard informed consent form because it is the current standard of practice and another objective of this study was to identify strategies for enhancing the exiting standards of informed consent. Furthermore, in order to isolate the influence of the social recommendation cue from highlights, which were used to indicate the passages that other participants had recommended, a third informed consent fo rm format with only highlights was used. This study also hypothesized that social recommendation cue s would differently cognitive processing among those with variable levels of n eed for cognition (NFC). Participants and Location This study was conducted through a partnership with the University of Florida TSI) biorepository project. The objective of their study was to build a biorepository of human tissue for future medical research. Tissues collected for the biorepository came from patients with varying demographic backgrounds who consent to donate their tissues which remain after medical procedures (surgery, biopsy, etc.). These tissues would otherwise be discarded after the procedures. Patients who were asked to donate their tissue to the biorepository had to undergo an informed consent process, and it was this consent process which served as a platform for this dissertation research.

PAGE 45

45 Research coordinators from the biorepository offered their resources and expert knowledge in the formulation of this study. Through this partnership, the informed consent form for the biorepository wa s used as the stimulus material for this study and arrangements were made for participants to be recruited from waiting rooms of surgical clinics that were in direct partnership with the biorepository. The use of both the real informed consent form and access to participants that were true candidates for the biorepository study was especially critical for making the study as true to life as possible. Caregivers were also included in the study because accumulat ed evidence has sho wn that caregivers play important roles in making medical decisions and are viewed as important audiences for patient education by medical providers (Prohaska & Glasser 1996). Although recruiting real biorepository participants and their caregivers would have been ideal, acute time and space limitations in the clinical setting did not make this a feasible option. Moreover, the informed consent process for the real biorepository study takes place at the end of each appointment, usually after patients have r eceived bad news, after extended wait times, and lengthy forms have been completed, so recruiting participants from waiting rooms was more favorable, because this reduced the probability of confounding factors such as severity of illnesses and levels of e xhaustion influencing the study Therefore participants for this study were patients and caregivers, 18 years of age and older, recruited from waiting rooms for surgical clinics that were associated with the biorepository project. Recruiting participant s associated with the CTSI biorepository was also beneficial because the biorepository used tablet computers to present their informed consent form to patients. Patients also used styluses to navigate and sign their digital consent forms,

PAGE 46

46 so the use of ta blets and styluses was already an established practice in this research setting. Thus, the inclusion elements needed to test the hypotheses for this study (social recommendation cues and highlights) was a smooth extension of an existing communication proce ss. Furthermore, research coordinators for the biorepository study reported that, after brief instructions on how to use the tablet and stylus, patients usually required little assistance in navigating and using the tablets and styluses, regardless of age, ethnicity, or sociodemographic background. Therefore, this study instructed participants on how to use the tablet and styl us es as a means to account for participant abilities to use and navigate tablet computers. This study was approved by both the medica l Institutional Review Board (IRB) and CTSI regulatory board at the University of Florida. This study was also exempt of informed consent by the medical IRB because of its low potential for risk and harm and because participants could not be told that they had been randomly assigned to view different consent formats until after the study had been completed in order to maintain the integrity of the study design. However, in place of a formal informed consent process, we opted to provide a form describing the study to participants during a post study debriefing process and gave them an opportunity to withdraw their data from the study Recruitment P rocedures Both recruitment and data collection took place in waiting rooms for surgical clinics. Waiting rooms se ated an average of 30 seats and seats were distributed around the waiting room to form roughly five different seating sections. These sections were divided by columns, planters, and other artificial dividers, so seating sections were somewhat intimate, but these sections could still be seen by receptionist and other staff

PAGE 47

47 members working at the front desks. Additionally, as per agreement with the biorepository and surgical clinics, this study had to make special efforts to not disturb clinical practices, wh ich meant that participants for this study could be called in to their appointment at any time. Therefore, participants could not be removed from waiting room s and take n to private spaces to conduct the study. Space was also limited in the waiting rooms, s o private seating areas for the study could not be a rranged Therefore, participants that agreed to take part in the study did so in their own seats, and other measures, described next, were taken to account for these limitations. This study aimed to recru it all patients in the waiting room and if patients were accompanied by one or more caregivers, data were only collected from either the patient or one of their caregivers. This was done to reduce the risk of two participants overhearing study instructions or information provided in the debriefing process (see Appendix A for details about the debriefing process) Patients were always the first to be asked to join the study because they were the primary targets of this study; however, the researcher enrolled caregivers in place of patients if a caregiver volunteered to As noted previously, this study aimed to recruit patient/caregiver units in the waiting room. However, data were collected from one participa nt at a time, so the ability to reach all patients was limited at times especially as the number of people in the waiting room and the rate at which they entered fluctuated throughout the day and week. Therefore, as a general rule, the next potential par ticipant that the researcher approached was the patient who most recently turned in his/her admissions paperwork which

PAGE 48

48 waiting had begun and that he/she also had the most wa it time remaining compared to other patients in the room. This also increased the probability that the patient was not around when study instructions were provided to the previous participant. The researcher also maintained a low voice throughout the study to avoid people overhearing, which was aided by the ambient noise of the waiting room and the degree of privacy provided by the different seating sections. In the event that the next potential participant seated him/herself in the same section as the cu rrent participant and was present to hear the debriefing process, the researcher approached the next patient, in a farther seating section, that handed in his/her admissions paperwork to the clinical staff The turnover rate in the waiting room was usuall y slow enough that the researcher could observe and identify the next potential participant, but on busier days, the clinical staff was helpful in identifying the next potential participant. S tudy P rocedures Using a standardized script (Appendix A ), patien ts and caregivers in surgical waiting rooms were approached by the primary researcher and asked if they were interested in providing feedback on how informed consent information for medical research was presented. If they responded favorably, they were imm ediately enrolled in the study and began their participation. The study began with the researcher explaining the study process to the participant, and it included a statement that was intended to increase topic importance and relevance for participants: lease remember that this consent form is for a real study, so you (or your loved this study is completely separate from the actual medical study, so you might be

PAGE 49

49 asked to go After this brief explanation, the researcher gave participants a study packet and guided the participa nts through the process of providing sociodemographic information and measuring perceived personal importance of learning about the study. While participants provided their responses, the researcher selected a randomly assigned consent format for the part icipant to view on the tablet. Consent formats were randomly assigned to each participant using a random number generator by Qualtrics. The application, which was available on the tablet, was pre set to generate one of the names of the consent formats at t he click of a button. Once participants completed the initial measures, the researcher demonstrated how to read the consent forms using the tablet and encouraged participants to highlight information that they thought was important using a stylus and an o nline highlighting application. The highlighting application was called Diigo and it allows people to highlight and share information on a webpage with a sharing tool (Figure 3 1 ). After the participant demonstrate d sufficient ability to utilize the feat ures on the tablet computer participants were instructed to carefully process information in the consent forms and the researcher re emphasized the importance of the topic: "Now, you need to know that any emphasized information is not the only information that's important for you to know. Reading this form is the only reliable way for you to learn all about the study. That's why you need to evaluate the form carefully -from beginning to end. When you're finished with the form, let me know, so you can s tart filling out the questionnaire." Participants w ere then instructed to take their time reading the consent form and to complete questionnaires provided in the packet when they were finished reading the consent form. Questionnaires measured the extent o f effort they put into processing the

PAGE 50

50 consent forms, n eed for cognition and their attitudes toward the consent p rocess and presentation format (Appendix C ). During this process, the researcher stepped back, remaining within hearing distance, and started p reparing materials for the next participant in order to allow the current participant to take his/her time. After the questionnaires were completed, the study ended and participants were given the opportunity to have an open discussion about the bioreposit ory study with the researchers The participants were then debriefed about the study and given an opportunity to withdraw their data f ro m the study. After the study was completed participant grouping according to level of need for cognition (NFC) low v s high was determined u sing a median split procedure employed in previous research on NFC (e.g. Cacioppo, Petty, & K ao, 1984; Evans & Petty, 2003) In this procedure NFC score s for each participant were calculated by adding up all items in the scale. Af ter the group median for NFC scores was identified, participants with NFC scores below the group median were assigned to the low NFC category and participants above the median were assigned to the high NFC category. The researcher also went back and delete d any passages that may have been highlighted by participants Study C onditions All study conditions were based on the informed consent form for the biorepository. In keeping with the standards of the local Institutional Review Board (IRB), the contents f or the informed consent form were structured like a legal document (e.g. full block format, bullets, section headings, etc.). All the contents were displayed as a continuous page through a web browser on an iPad but as a reference, the document was 8 pag es long in its origin al, IRB approved paper format (s ee Appendix B

PAGE 51

51 for the contents of the consent form ). All instructions and procedures were the same, except for the instructions provided in relation to existing highlights and/or the ability to highlight as detailed below. Standard C ondition with No C ues Participants in the standard condition were simply asked to read the standard informed consent form (Figure 3 2) T hey were asked to highlight information that they thought was important, and were given t he following instructions by the researcher : "To help you process the consent form, here's an online version of the form. Using the digital highlighter, highlight information that you think is important for you and anyone else to know." S ocial Recommendat ion Cue C ondition Social recommendation cues for this condition revolved around passages of the informed consent form that previous participants had highlighted as recommendations for others to read The cue that conveyed the social presence of others was a small icon that appeared to the left of every highlighted passage and denoted the number of other participants that had highl ighted each passage (Figure 3 2) These icons were generated using Diigo, the online highlighting application, which also allows people to share information they highlighted with others. Icons appear when more than one person highlights a passage on a shared webpage. In order to display recommendation cues that were consistent from one participant to another, the highlights, icon s, and the numbers denoted in each icon were hardcoded into the consent form for this condition Therefore, all participants in this condition saw the same social recommendation cues and were given these instructions: "To help you process the consent form, here's an online collaboration system that allows you to see what other patients and caregivers, like you, thought was important for you to know. Using the digital highlighter, highlight information that

PAGE 52

52 you think is important for other patients to know. If it has been highlighted by other people already, just re Additionally, the highlighted passages were based on findings from a previous study that asked stakeholders of another biorepository study to highli ght information they perceived to be valuable (Beskow et al., 2010). This lead to 17.3 % (435 words out of 2,509 words) of the form being highlighted (Appendix C ). The numbers denoted in the ch gave the count for how many of their participants highlighted each passage. Highlights Only C ondition The informed consent form for this condition was nearly identical to the social recommendation cue condition but it did not include the icon that indic ated the number of other participants who had highlighted the passages (Figure 3 2) Additionally, participants in this condition were given the same instructions as those in the standard condition: "To help you process the consent form, here's an online version of the form. Using the digital highlighter, highlight information that you think is important for you and anyone else to know." The purpose of this condition was to help isolate the influence of the social recommendation cues from the highlights a nd because, according to ELM, enhancing ability or readability of the form could also enhance probability of people taking the central path to information processing. informed consent forms also suggested that highlighting key passages could make consent forms more manageable for people to process, which in turn would encourage more careful evaluations of information in the forms.

PAGE 53

53 Measures Independent Variable Need for cognition Need for cognition (NFC) is the dispositional characteristic of an individual and their tendency to enjoy engaging in effortful analytical activities (Cacioppo, Petty, & Kao, 1984). The n eed for cognition s cale will be used to distinguish the continuum of those who naturally enjoy though tful thinking (high in NFC) and those who do not (low in NC). Items for this scale include, "Thinking is not my idea of fun," and "I would prefer complex to simple problems." Each item is rated on a 5 point unipolar scale with end points n ot like v (Cohen, Stotland, & Wolfe, 1954) and later advanced by Cacioppo and Petty (1982; 1984), this scale h as consistently achieved high internal consistencies and its validity has been supported through its modest correlations with verbal intelligence, ACT scores and grade point averages (Cacioppo et al., 1996). This scale traditionally has 18 items but because time was limited in the clinical setting, nine of the most strongly correlated items (19 82). Abbreviated versions of the scale have been used in other studies ( Smith & Levin, 1996 ). This scale is often measured several days in advance of the experimental procedures in order to minimize its p otential confounding influence (Cacioppo Petty, & Morris, 1983; Evans & Petty, 2003); however, this study was limited by participant availability, so NFC was measured at the end of the study and group membership (high or low NFC) was determi ned post hoc Group membership was determined u sing a median split procedure employed in previous research on need for cognition (e.g. Cacioppo, Petty, &

PAGE 54

54 Kao, 1984; Evans & Petty, 2003) See Appendix C for the full NFC scale and all measures used in this s tudy. Sociodemographic information Age, sex, ethnicity/race, education, and employment status were also collected to help account for differences between conditions and to examine variations in outcomes based on age, sex, and other demographic and social c haracteristics and to control for confounding variables Sociodemographic indicators were based on those used by the biorepository study. Manipulation C hecks Topic importance Topic importance was checked before participants were exposed to study condition s to control for potential differences in perceived topic importance. Topic importance is "a person's perception of the amount of personal importance he or she attaches to an attitude" (Wegener, et al., 1995). Perceived importance of evaluating the consent form was measured using two items which were adapted from Boninger, Krosnick, and Berent (1993) and Wegener et al., (1995). These Items include d now, how important is it for you to learn as much as you can about the study from this consent form? (Not important at all personally care about evaluating this informed consent form? (Not at all A great deal)." Responses will be given through a 7 point bipolar scale. Motivation to process information Bas to process information was evaluated across conditions. Motivation to process consent information was checked using four items measured with 5 point Likert scales. Items

PAGE 55

55 included, more I read, the less motivated I became about learning about the study (reverse coded) Ability to process information Ability to process information was checked using four items measured with 5 point Likert easy easy to fin Dependent Variables Cognitive e ffort According to Wegener, Downing, Krosnick, and Petty (1995 ) and their review of techniques for studying attitude strength cognitive effort can be measured through t hought listing exe rcises or s elf reported attitude measures Cognitive effort is often measured using a thought listing exercise, a technique in which participants are asked to list their thoughts after a given task but because time is restricted for this study, fo ur item s adapted from the work of Cacioppo, Petty, and Morris (1983) were used to point bipolar scale. Endpoints in you tried to do your best to ma ke sense out of this See Appendix C for the full cognitive effort scale and all measures used in this study.

PAGE 56

56 Attitude toward overall exper ience Participant attitudes toward their experience processing the consent form. The given on 9 point semantic differential attitude scales with endpoint s labeled: unfavorab le favorable, unpleasant pleasant, boring interesting, bad good, informative uninformative, and negative positive. This item was adapted from a measurement used in a similar experiment conducted by Evans and Petty (2003), in which participant attitudes tow ard the design and layout of the presentation and the content of the presentation were examined. Attitude toward presentation format Participant attitudes toward the presentation format were also measured using items adapted from the study conducted by E vans and Petty (2003) The item read, given on 9 point semantic differential attitude scales with endpoints labeled: unfavorable favorable, unpleasant pleasant, boring intere sting, bad good, informative uninformative, and negative positive. Pretest A pretest explored whether the social recommendation cues used for this study influenced cognitive effort expended in processing information. This pretest was conducted using an onl ine questionnaire that was hosted by Qualtrics, a third party online survey vendor approved by the university. Students ( N = 29) in an introductory advertising course were sent a link to the study materials through an approved listserv. When they clicked on the link, an informed consent form appeared which they had to

PAGE 57

57 read and accept before proceeding to the study (the consent form and study design were approved by the university's Institutional Review Board). Once the study began, students were asked to message expressed the importance and relevance of the topic to their academic career and urged students to read the honor code carefully. Students were randomly assigned to view different versio ns of the honor code through the survey program's random survey generator. The first version was a control version and had not highlights. The second version of the honor code had the social recommendation cues which included passages highlighted with icon s indicating the number of students who had highlighted each passage. After students finished reading the honor code, they were automatically prompted to complete a questionnaire that assessed the amount of effort they expended in processing the informatio n. Cognitive effort was measured using a 3 item scale, which was based on previous works by Wegener, Downing, Krosnick, and Petty's (1995). Responses were given through 7 point bipolar scales and included the items, "How much effort did you put into evalu ating the highlighted information?", "To what extent does the highlighted information motivate you to read further?", and "How much effort did you put into evaluating the whole document?" (Cronbach's alpha = .81, M = 11.63, SD = 4.0 3 ). An independent samp le t test indicated that students viewing the honor code with student highlights exercised significantly greater cognitive effort ( M = 4.30, SD = 1.41) than students who viewed the control honor code ( M = 3.06, SD =1.37), t (27) = 2.313, p = .03.

PAGE 58

58 Statistic al Analysis In line with prior empirical research on cognitive effort and need for cognition (e.g. Cacioppo, Petty & Morris, 1983; Haugtvedt, Petty & Cacioppo, 1992; Zhang, 1996), a two way analysis of variance (ANOVA) was performed to analyze the data Given that the current study aimed to test the effect of two factors need for cognition (low vs high) and study conditions (standard condition with no cues, social recommendation cue condition, highlights only condition) on cognitive effort as the dep endent measure, a two way ANOVA is more appropriate than a one way ANOVA (Hair, Black, Babin, Anderson & Tatham, 2010). As recommended by Hair et al., (2010), statistical significance was determined at p value of < .05.

PAGE 59

59 Figure 3 1. Diigo tool bar for iPad

PAGE 60

60 A B C Figure 3 2. Study conditions A) standard consent form, B) consent form with social recommendation cues and C) consent form with highlights only.

PAGE 61

61 CHAPTER 4 RESULTS Study Results Data were collected over a period of three months, st arting in May 2012 and ending in August 2012. Data were collected during all days of the we ek and times of the day spanning evenly from opening to closing of surgical clinics ( 9am to 5pm). The sample consisted of 113 usable responses after all incomplete questionnaires were removed. Participant age ranged from 18 to 85 years, M = 53.3, SD =15.8. Of the sample, 61.1% ( n = 69) were female and 35.4% ( n = 40) were male, with 3.5% ( n = 4) not reporting their gender. Participant roles, defined as either patient s or caregivers, were spread out nearly evenly with 50.4% ( n = 57) being caregivers and 46% ( n = 54) being patients. Similar to gender reporting, 3.5% ( n = 4) did not report their roles. Most participants had completed one year of college or more (72.6%, n = 82) and 24.4% ( n = 28) had completed high school (or completed their GED) or less. See Table 4 1. Scale Reliability Analysis Need for cognition Need for cognition (NFC) was measured using nine items from Cacioppo and n eed for cognition s cale (1 982). Endpoints of the 5 n ot like me v ery item correlations ranged between r = resulted from the deletion of any scale items. Points were assigned to each response by centering neutral responses on zero and successively assigning positive figures to

PAGE 62

62 composite scores were then calculated by adding up all the points and a median split procedure was used to assign participants to low and high NFC groups. Cognitive e ffort Four items adapted from th e work of Cacioppo, Petty, and Morris (1983) were alpha was .87 with inter item correlations between .67 and .87. This may be because the item was difficult to under stand or because there were too many distractions in the study environment. As recommended by Hair et al., (2010), this item was deleted based on its poor reliability and three items were used for the final analysis. Points were assigned to each response b y assigning zero to neutral responses and sequentially by adding up all the points. A ttitude toward overall experience Participant attitudes toward their experience processing the consent form were Responses were given on 9 point semantic differential att itude scales with endpoints labeled: unfavorable favorable, unpleasant pleasant, boring interesting, bad good, informative uninformative, and negative positive. A composite score was created by adding up responses from each semantic differential scale. Cro measure was .95, inter item correlations for the items were between r =.60 and r =.94.

PAGE 63

63 Attitude toward presentation format Participant attitudes toward the presentation format were measured using an responses were given on 9 point semantic differential attitude scales with endpoints labeled: unfavorable favorable, unpleasant pleasant, boring interesting, bad good, informative uninformative, and negative positive. A composite score was created by measure was .95, inter item correlations for the items were between r =.60 and r =.94. Topic importance Topic importance was me for you to learn as much as you can about the study from this consent form? (Not important at all evaluating this informed consent form? (Not at all A great deal)." Responses will be given through a 7 point bipolar scale. A composite score was created by adding up item correlation for the item was r = .6 6 Motivation to process information Motivation to process consent information was checked using four items measured with 5 less motivated I became responses and sequentially assigning positive points to r esponses that leaned toward

PAGE 64

64 for this measure was 70 inter item correlations for the items were between r = .24 and r = .67. All items were used for the final analysis because scale reliability would not have improved by removing any item. Ability to process information Ability to process information was checked using four items measured wit h 5 se by assigning zero to neutral responses and sequentially assigning positive points to l the item correlations for the items were between r = .33 and r = .81. All items were used for the final analysis because scale reliability would not have improved by removing any item. Manipulation C h ecks In order to test how each condition i nfluenced motivation and ability to process information two separate 3 x 2 analysis of variance (ANOVA) was conducted. The independent variables were study conditions ( standard consent form with no cues consent form with social recommendation cue s and consent forms with high lights only ) and need for cognition (low and high) and the dependent variable was either the measure for motivation to process information or ability to process information The 3 x 2 ANOVA for motivation revealed a significant main effect for study con dition s, F( 2, 107) = 6.01, p = .003. The analysis also revealed a significant main

PAGE 65

65 effect for NFC, F( 1, 107) = 5.51, p = .021, but did not reveal an interaction between study conditions and NF C, F( 2, 107) = .44, p = .64 (see Table 4 Significant Difference (HSD) post hoc comparisons of the three groups indicated that the participants in the social recommendation cue condition ( M = 3.39, SD = 1.98) were significantly more mot ivated than those in the highlights only condition ( M = 1.54, SD = 2.97), p = .004. However, motivation in the standard condition did not differ significantly from the other two treatments. Means, standard deviations, sample sizes for each group are provi ded in Table 4 3. The 3 x 2 ANOVA for ability processing information, failed to reveal a significant main effect for study condition s, F( 2, 107) = 1.10, p = .34 However, a significant main effect for NFC was revealed, F( 1, 107) = 8.90, p = 004, and th e interaction between study condition s and NFC was not significant, F( 2, 107) = 1.81, p = 17. Mean ability scores for standard highlights only and social recommendation cue condition s were, 2.58 ( SD = 2.11), 1.59 ( SD = 3.44), and 2.02 ( SD = 3.062), re spectively. Positive scores indicated that perceive ability to process consent information was generally high across treatments. A reexamination of NFC indicated that those with high NFC ( M = 1.25, SD = 2.91) significantly perceived greater ability than t hose with low NFC ( M = 2.88, SD = 2.73). Notably, these mean scores also were positive Combining these findings with the previous manipul ation check for motivation, differences in cognitive effort between conditions are likely attributable to motivation to process information. See Table 4 4 and 4 5. Finally, another 3 x 2 ANOVA was conducted to test whether or not topic importance was consistent across all study conditions The independent variables were

PAGE 66

66 study conditions ( standard, recommendation cue, an d highlights only ) and N FC (low and high) and the dependent variable was perceived importance of learning about the study. The analysis revealed no significant main effect for study condition s, F( 2, 107) = 0.21, p = 0.81, or for NFC, F( 1, 107) = 0.92, p = 0.34. Interaction between treatments and NFC also was not significant, F( 2, 107) = 0.13, p = 0.87. Overall, participants rated learning about the study as important M = 5.13, SD = 1.75 (see Table 4 6 and 4 7). According to the elaboration likelihood mo del individual perceptions of topic importance is directly related to motivation to process information, so this lack of significant difference in group means suggest that treatment effects on cognitive effort can be attributed to motivation generated by the study treatments, especially for the social recommendation cue and highlights only condition s. Therefore, the manipulation yielded the correct conditions for testing study hypotheses.

PAGE 67

67 Hypothes i s Testing Study hypotheses were simultaneously analyzed u sing a 2 way factorial analysis of variance (ANOVA). The independent variables were study conditions (standard, social recommendation cue, and highlights only) and NFC (low and high) and the dependent variable was cognitive effort expended in processing co nsent information H1: Consent forms with social recommendation cues will increase cognitive processing compared to consent forms with no cues H2: Consent forms with social recommendation cues will increase cognitive processing compared to consent fo rms with only highlighted pass ages H3: Consent forms with only highlighted passages will increase cognitive processing compared to consent forms with no cues. The ANOVA revealed a significant main effect for study conditions, F( 2, 107) = 3.23, p = hoc comparisons of the three groups indicated that those in the social recommendation cue condition ( M = 4.53, SD = 3.52) expended greater cognitive effort than the highlights only conditions ( M = 2.35, SD = 4.13), p = 05. However the post hoc comparisons indicated that cognitive effort in the standard condition ( M = 4.05, SD = 4.18) did not differ significantly from the social recommendation cue condition, p = 0.86, or the highlights only condition, p = 0.15, and mean cognitive e ffort was above the highlights only condition instead of below Thus, H2 was supported but H1 and H3 was not supported H4: NFC is positively related to cognitive effort that participants expended in processing consent information, with those with low NFC exercising less cognitive effort than those with high NFC. The analysis revealed a significant main effect for NFC, F( 1, 107) = 4.02, p = 0.05. Those with high NFC ( M = 4.33, SD = 3.38) reported greater effort expended in processing consent information t han those with low NFC ( M = 2.96, SD = 4.52). No

PAGE 68

68 interaction between study conditions and NFC were revealed, F( 2, 107) = 0.22, p = 0.81 (see Table 4 8). Therefore, those with high NFC expended greater cognitive effort than those with low NFC, regardless o f study conditions, and H4 was supported. H5: Social recommendation cues will increase cognitive processing of informed consent information among those with low NFC. The graphical display of interaction effects of cognitive effort across study condition s by NFC is provided in Figure 4 1. Means and standard deviations for each group mean for cognitive effort are also provided in Table 4 9. The mean cognitive effort for those with low NFC in the social recommendation cue condition was 4.13, SD = 4.03. T his was greater than both the means for those in the highlights only and standard conditions ( M = 1.67, SD = 4.10 and M = 2.73, SD = 5.48, respectively). Because both main effects were significant, H 5 was supported. Importantly, the ANOVA did not reveal a n interaction between study conditions and NFC. Therefore, the higher level of cognitive effort in the social recommendation cue condition appears to be due to the motivational influence of social recommendations and independent of inherent motivational te ndencies related to NFC.

PAGE 69

69 Figure 4 1. Interaction effects of cognitive effort by n eed for cognition H6: Social recommendation cues will have no impact on cognitive processing among those with high NFC. Results from the previous ANOVA demonstrated tha t even those with high NFC were more likely to exercise higher effort when they were exposed to social recommendation cues than those who were not ; therefore, H6 was not supported Exploration of Research Questions Finally, two separate 3 x 2 analyses of variance (ANOVA) were conducted to answer the two research question s The independent variables were study conditions (standard, social recommendation cue, and highlights only) and NFC (low and high) and the dependent variable was either the measures asses sing participant attitudes toward their overall experience learning about the study and their attitudes toward the presentation format of the consent form RQ 1 : Do people exposed to recommendations made by other people evaluate their overall experience di fferently than those who are not?

PAGE 70

70 The ANOVA for overall experience did not reveal a main effect for study conditions F( 2, 107) = 2.71, p = 0.07. However, th e analysis revealed a significant main effect for NFC, F( 1, 107) = 5.59, p = 0.02, but did not re veal an interaction between study conditions and NFC, F( 2, 107) = 0.34, p = 0.71 (see Table 4 10 ). Overall, those with high NFC rated their experience more favorably ( M = 6.47, SD = 3.24) compared to those with low NFC ( M = 4.95, SD = 4.26). RQ 2 : How do participants evaluate the use of social recommendation systems to help communicate informed consent information? T he ANOVA for overall attitude toward the presentation format did not reveal a main effect for study condition s, F( 2, 107) = 1.37, p = 0.26, o r for NFC, F( 1, 107) = 3.49, p = 0.06. The interaction between study conditions and NFC was also not significant, F( 2, 107) = 0.32, p = 0.73 (see Table 4 1 1 ).

PAGE 71

71 Table 4 1. Participant characteristics Sociodemographics ( N = 113) n % Age, years ( M = 5 3 ) < 5 3 54 47.8 3 59 52.2 Sex Female 69 61.1 Male 40 35.4 Unreported 4 3.5 Ethnicity Hispanic 8 7.1 Non Hispanic 103 91.2 Unknown 2 1.8 Race White 95 84.1 Black or African 17 15 American Asian Japanese 0 0 Asian Non Japanese 0 0 Hawaiian or other Pacific Islander 1 0.9 Native American 0 0 Education High school or less 28 24.5 Some college or more 82 72.6 Unreported 0 0 Role Patient 52 46 Caregiver 57 50.4 Unreported 4 3.5

PAGE 72

72 Table 4 2. Between subject effects for group differences in motivation across study conditions by need for cognition Source df SS MS F p Corrected Model 5 103.78a 20.756 3.491 0.006 Intercept 1 706.62 706.62 118.84 0.0 0 Study Condition 2 71.50 35.75 6.01 0.003 Need for Cognition 1 32.78 32.78 5.51 0.02 Study Condition x Need for Cognition 2 5.28 2.64 0.44 0.64 a. R Squared = .140 (Adjusted R Squared = .100)

PAGE 73

73 Table 4 3. Description statistics of motivation for study conditions by need for cognition Study Condition Need for Cognition M SE n Standard Low 2.27 2.55 15 High 2.96 2.31 23 Highlights only Low 0.67 3.38 18 High 2.37 2.31 19 Social recommendation Low 3.04 2.20 23 cue High 3.93 1.49 15 Total 113

PAGE 74

74 Table 4 4. Between subject effects for group differences in perceived ability across study conditions by need for cognition Source df SS MS F p Corrected Model 5 117.25 23.45 2.9865 0.01 5 Intercept 1 470.58 470.58 59.93 0.00 Study Condition 2 17.20 8.60 1.10 0. 3 4 Need for Cognition 1 69.89 69.89 8.90 0.00 Study Condition x Need for Cognition 2 28.35 14.17 1.80 0. 1 7 a. R Squared = .122 (Adjusted R Squared = .081)

PAGE 75

75 Table 4 5. Description statistics of perceived ability for study by need for co gnition Study Condition Need for Cognition M SD n Standard Low 2.20 1.66 15 High 2.83 2.37 23 Highlights only Low 0.06 3.65 18 High 3.05 2.55 19 Social recommendation Low 1.57 2.69 23 cue High 2.73 3.53 15 Total 113

PAGE 76

76 Table 4 6. Be tween subject effects for group differences in topic importance across study conditions by need for cognition Source df SS MS F p Corrected Model 5 5.64 1.13 0.357 0.88 Intercept 1 2851.79 2851.79 903.40 0.00 Study Condition 2 1.35 0.67 0.21 0.8 1 Need for Cognition 1 2.90 2.90 0.92 0.34 Study Condition x Need for Cognition 2 0.84 0.42 0.13 0.87 a. R Squared = .017 (Adjusted R Squared = .030)

PAGE 77

77 Table 4 7. Description statistics of topic importance for study conditions by need for cog nition Study Condition Need for C ognition M SD n Standard Low 5.07 1.75 15 High 4.91 1.86 23 Highlights only Low 5.41 1.54 18 High 4.84 1.92 19 Social recommendation Low 5.39 1.47 23 cue High 5.13 2.13 15 Total 113

PAGE 78

78 Table 4 8. Betw een subject effects for group differences in cognitive effort across study conditions by need for cognition Source df SS MS F p Corrected Model 5 166.44 a 33.29 2.15 0.06 Intercept 1 1417.33 1417.33 91.74 0.00 Study Condition 2 99.78 49.89 3.23 0 .04 Need for Cognition 1 62.09 62.09 4.02 0.05 Study Condition x Need for Cognition 2 6.69 3.35 0.22 0.81 a. R Squared = .140 (Adjusted R Squared = .100)

PAGE 79

79 Table 4 9. Description statistics of cognitive effort for study conditions by need for c ognition Study Condition Need for Cognition M SD N Standard Low 2.73 5.48 15 High 4.91 2.89 23 Highlights only Low 1.67 4.10 18 High 3.00 4.16 19 Social recommendation cue Low 4.13 4.03 23 High 5.13 2.59 15 Total 113

PAGE 80

80 Table 4 10 Be tween subject effects for group differences in overall experience across study conditions by need for cognition Source df SS MS F p Corrected Model 5 151.59 30.32 2.16 0.06 Intercept 1 3591.76 3591.76 255.98 0.00 Study Condition 2 76.08 38.04 2. 71 0.07 Need for Cognition 1 78.44 78.44 5.59 0.02 Study Condition x Need for Cognition 2 9.57 4.78 0.34 0.71 a. R Squared = .092 (Adjusted R Squared = .049)

PAGE 81

81 Table 4 11 Between subject effects for group differences in attitude toward presenta tion format across study conditions by need for cognition Source df SS MS F p Corrected Model 5 96.41 19.28 1.20 0.32 Intercept 1 3379.48 3379.48 210.01 0.00 Study Condition 2 44.03 22.02 1.37 0.26 Need for Cognition 1 56.10 56.10 3.49 0.06 St udy Condition x Need for Cognition 2 10.21 5.10 0.32 0.73 a. R Squared = .053 (Adjusted R Squared = .009)

PAGE 82

82 CHAPTER 5 DISCUSSION The purpose of this study was to examine how social recommendation cues influenced cognitive processing of informed consent information for medical research. Contrary to the normative view that social cues promote peripheral processing, results from this study demonstrated that social recommendation cues can significantly increase cognitive effort expended in processing information. What is more, the effect of social recommendation cues was one mediated by increased motivation rather than a change in perceived ability. This main effect, of social recommendation cues on motivation to process information and cognitive eff ort, was independent of both need for cognition (NFC) and the passage highlights, which were predicted to enhance readability of the form. The passage highlights were predicted to make the consent form look easier to read and process, thereby increasing pe rceived ability. However, participants exposed to highlights both in the social recommendation cue and highlights only conditions did not report that the highlights improved their ability to process information. Furthermore, in comparison to the social recommendation cues, informed consent forms with only passage highlights decreased both motivation to process information and cognitive effort. The fact that highlights did not affect perceived ability but, instead, decreased motivation, was not an expect ed finding. A possible explanation is that the highlights inadvertently took on the form of a social cue that initiated peripheral processing. In the absence of an attribution for the highlighting, the highlights could have been attributed to the medical researchers who authored the consent form. Therefore, it appears as if participants relied on source

PAGE 83

83 credibility as part of their peripheral processing, which motivated participants to expend low cognitive effort. This is plausible given a significant bod y of research that has demonstrated how social cues based on source expertise can initiate peripheral processing (Petty & Cacioppo,1981; Petty & Wegener, 1999; Perloff, 2010; Baran & Davis, 2009). The influence of social cues on cognitive processing rela tive to the standard consent form remains unclear, because mean cognitive effort expended by those in the standard condition did not differ significantly from either the means of the social recommendation cue or highlights only conditions, as a result of t he data variability for the pooled sample of the standard group. One possible explanation for why the standard consent form did not have a clear examines the extremes within the elaboration continuum (Perloff, 2010; Petty et al., 1993). In other words, ELM can help explain what does and does not work to significantly influence cognitive effort, but not what occurs in between. Another possible explanation is that the normal var iability from confounding factors that is expected when people encounter the standard consent form was not controlled for in the standard condition. Nevertheless, need for cognition (NFC) was a significant predictor of cognitive effort expended in processi ng consent information across all conditions, an effect which was especially pronounced in the standard condition. Conclusion Until this present study, no studies had examined the application of social recommendation cues as an independent factor meant to increase cognitive processing. Therefore, an important contribution of this study was that it revealed mechanisms

PAGE 84

84 through which social recommendation cues influence cognitive effort expended in processing issue relevant information. That is, social recom mendation cues influence motivation to process information. Moreover, this relationship between social recommendation cues and its influence on motivation appears to be moderated by source credibility and social identification with messages sources. Socia l recommendation cues, contrary to their (through social identification, such as homophily or in group/out group source comparisons) or relegate processing to the peripher al route, where reliance upon expert sources is apparent. "appropriate heuristic" or an individual's perceived match between the message sources and forums (e.g. doctors are credible on professional websites, but less so on discussion boards) is a mediator of social credibility and influence on health messages (2010). Research by Mackie, Gastardo Conaco, and Skelly (1992) also indicates differential cognitive processing of messages f rom in group and out group sources. Sources for this study group. Therefore, social identificat ion with message sources even as social cues appear to a ffect the valence of social recommenda tion cues on motivation to engage with message contents which then influence strategies for processing information I mplications for Mass Communication R esearch While this study demonstrated how social recommendation cues can positively influence cognitiv e processing, similar findings have also been demonstrated in other areas of social science research (Soll & Larrick, 2009 )

PAGE 85

85 contributions is that it helps bridge the gap between ELM and the other areas of research that have ide ntified social cues as salient forces that influence attitude change. A key factor that seems to bridge this gap is the influence of the social attributes of message and cue sources on motivation. These social attributes are often defined in terms of sourc e credibility in ELM literature, but this dimension seems to be incomplete. Hu and Sundar (2010) measured both source credibility and homophily in their study of the effects of online health sources on behavioral intentions. These findings were based on ELM, and provided a more robust explanation of the influence of social attributes of message sources on attitudes. Despite this, their outcomes were not entirely explained either; therefore, as their inability to capture the social attributes of message so urces highlights, one of the greatest challenges still facing mass communication research is the very task of conceptualizing the social attributes of this Theoretical I mplications One criticism of ELM is that it does not clearly indicate h ow messages should be formulated because it does not clearly distinguish between features that are peripheral and central, and some researchers do not agree that there is a dual route to persuasion (Perloff, 2006; Petty et al., 1993). In fact, social cues, such as the icon indicating the number of people who highlighted the passages, are generally categorized as peripheral features in ELM experiments (Evens & Petty, 2003). However, this study demonstrated that social recommendation cues could motivate grea ter cognitive effort and influence people to take the central processing route. Therefore, it appears that peripheral and central features of a message can act concomitantly to motivate cognitive effort. Those with high NFC have the disposition that inhere ntly drives them

PAGE 86

86 through to central processing, regardless of peripheral features, while those with low NFC depend on peripheral features (such as the social recommendation cues) to determine whether or not they should exercise greater effort to understand information. By identifying these characteristics in target populations, content can be modified to promote central processing, and therefore better decision making, universally. Implications for Informed Consent a nd Medical Communication R esearch The no rmative view of informed consent is that it is a neutral process, free of undue forces of influence, formulated to help patients make voluntary and informed decisions about joining medical research studies. However, results from this study demonstrated how subtle changes in the communication process could influence how influences how much effort par ticipants expend in processing consent information. Recommendations of other participants, which had a favorable impact on cognitive effort, are commonly unavailable in informed consent processes, while highlights by medical researchers, which had an unfa vorable impact on cognitive effort, are commonplace in the informed consent process (whether written or verbal). Because medical researchers and physicians customarily summarize informed consent forms for subjects and patients, the impact that these highli ghts and summaries have on patient decisions also warrants further research. Additionally, this study demonstrated how motivational factors significantly influenced information processing; therefore, efforts to improve informed consent should also account for how participants are motivated to process information. Focusing on motivation will be a novel approach to informed consent research because efforts to

PAGE 87

87 improve informed consent traditionally focuses on ability related factors that inhibit informed con sents, such as simplifying forms (Paris, Chaves, & Cornu, 2007) and incorporating audiovisual aids (Houts et al., 2001) or interactive computer games (Bickmore, Pfeifer, & Paasche Orlow, 2009). Although enhancing perceived ability is important, results fr om this study suggest that examining how different communication strategies for informed consent affect motivation may reveal more underlying truths about why certain strategies prompt greater cognitive effort, allowing for a more truly informed consent th an others. Observing how NFC is predictive of how people make judgments as well as the stability those judgments, NFC could be explored as a proxy for evaluating informed consent and for developing strategies for communicating informed consent. Additiona lly, the abbreviated NFC scale with 9 items reliably and efficiently distinguished between high and low NFC. This may be particularly helpful in the clinical setting, due to the limited time and resources available to evaluate how patients respond to medic al communication processes. In conclusion, examining dimensions of the informed consent process, such as source credibility, as well as outcome variables, such as cognitive effort, is an essential strategy for understanding and improving the informed conse nt process. Limitations and Future Work Because this study did not anticipate the influence of highlights on cognitive effort and the social cue it may have embodied, future studies should examine how source credibility and audience identification with mes sage sources influence the relationship between social recommendation cues and cognitive effort. Additionally, this study relied upon an important assumption: the cognitive effort expended in processing consent

PAGE 88

88 information is indicative of acquisition of relevant information and informed decision making for joining medical research studies. Therefore, future studies must examine whether this is, in fact, true and applicable to the informed consent process. By evaluating whether cognitive effort effectivel y correlates with participant knowledge of medical studies as well as their long term attitudes toward studies, confirmation of the robustness of this analysis can help guide its usage in future work. Additional limitations of this study were the limited s ample size and the fact that participants were asked to complete the study in waiting rooms, constrained by the space and time limitations of the clinical setting. In order to provide as realistic an experience as possible, the study environment was not co ntrolled so there could have been confounding variables that have not been accounted for in this study (though testing conditions were considered to be consistent across subjects). Additionally, participants were not real patients undergoing an informed co nsent process, so the direct application of results from this study for informed consent is not clear, especially because during real consent processes, patients also have to account for the severity of their illnesses and levels of possible risks associat ed in participating in medical research studies. Applying this model to real patient and subject informed consent processes would provide insight into the ability to garner truly informed decisions. Finally, findings from this study related to the use of s ocial recommendation systems for communicating health information should also be applied cautiously, because it may lack generalizability. Additionally, with the application of any new technique, unforeseen confounders could arise. For example, this stud y did not give participants free reign to recommend highlights, but instead, highlights were based on

PAGE 89

89 published research by Beskow et al. (2010) and passages that stakeholders from the biorepository had highlighted. These highlights also included those tha t were made by members of the Institutional Review Board and medical investigators, and not just participants. Therefore, future studies should examine content highlighted by participants alone, in comparison to that of the investigators, and assess how su ch a change might influence the informed consent process. Application across varied contexts will also provide more insight into the impact of social recommendation cues on decision making.

PAGE 90

90 APPENDIX A RESEARCHER SCRIPT AND INSTRUCTIONS FOR PARTICIPANT R ECRUITMENT AND ENROLLMENT 1. Introduction Researcher: Introduce yourself to the patient and family. E.g. Hello, my name is informed consent form for a real medical study. Do yo u think you could help us out by Patient/ caregiver: Yes/ No a. "No" response excluded from this study b. "Yes" response Researcher : "Okay great, but please remember that this consent form is for a real study, so you (or your loved one) could be is completely separate from the actual medical study, so you might be asked to go over this same information again. Also, becaus medical study, that impacts real lives, please remember that your 2. Participants fill out measures in study packet Researcher : Give participant a study packet and asked him/her to provide information about his/her gend er, age, race, ethnicity, role (patient or caregiver), and perceived importance of the topic of informed consent, and highest level of education they had completed (see Appendix C for all study measures). Ask him/her to let you know when he/she is finished 3. Select and explain the informed consent format Instructions for researchers: a. Step 1: Pull up randomly assigned consent format: A) Standard, B) Highlights only, C) Social recommendation cue. (Note: A random number generator application by Qualtircs was p re set on the tablet to generate the name of one of the formats at the click of a button.) b. Step 2: Explain and demonstrate presentation format to the participant: i. If Standard or Highlights only consent format, say, "To help you process the consent form, he re's an online version of the form. Using the digital highlighter, highlight information that you think is important for you and anyone else to know." ii. If Social recommendation cue, say, "To help you process the consent form, here's an online collaboration system that allows you to see what other patients and caregivers, like you, thought was important for you to know. Using the digital highlighter, highlight information that you think is important for other patients to know. If it has been highlighted by ot her people already, just re highlight the c. Step 3: Say all participants, "Now, you need to know that any emphasized information is not the only information that's important for you to know. Reading this form is the only reli able way for you to learn all about the

PAGE 91

91 study. That's why you need to evaluate the form carefully -from beginning to end. When you're finished with the form, let me know, so you can start filling out the questionnaire." d. Step 4: Give the tablet and styl us to the participant, make sure he/she knows how to use the devices by demonstrating how to scroll, zoom, and highlight passages on the tablet. Observe him/her for a few moments to make sure he/she is scrolling at a reasonable rate and appears to be readi ng the consent form comfortably on the tablet. Step back and work to "look busy" in order to give him/her space and time to process information at his/her own pace. For example, you can prepare study materials for the next participant. Note: If other careg ivers/ family members/ friends, etc., are around, say, "Until the questionnaires are finished, please give him/her time to look over the consent form him/herself. Then, everyone will have a chance to discuss the study details as a group." (Also give paper copies of the consent form to caregivers/ family members so they have something to review). 4. Participants fill out final measures in packet Researcher: After a participant signal to you that he/she has finish evaluating consent information, approach them an d say, "You'll have a chance to ask any questions you have about the study. But first, we want to learn about your reaction to the consent process." And ask him/her to complete the remaining measures in the study packet. 5. Debriefing Once the participant co mpletes the final measures and the study ends study ends, explain to him/her that they viewed one of three experimental formats for presenting informed consent information. Give them the form that describes the study and give him /her an opportunity to ask further questions about the study and the opportunity to be removed from the study.

PAGE 92

92 APPENDIX B BIOREPOSITORY INFORMED CONSENT FORM WITH SOCIAL RECOMMENDATION CUES

PAGE 93

93

PAGE 94

94

PAGE 95

95

PAGE 96

96 APPENDIX C STUDY INSTRUMENTS Topic importance

PAGE 97

97 Cognitive effort Attitude toward overall experience and presentation format Motivation to process information

PAGE 98

98 Ability to process information Need for cognition

PAGE 99

99 LIST OF REFERENCES Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes 50, 179 211. Areni, C.S., Ferrell, M.E., & Wilcox, J.B. (2000). The persuasive impact of reported group opinions on individual low vs. high n eed for cognition : Rationalization vs. biased elaboration? Psychology and Marketing 17(10), 855 875. Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgments. Groups, Leadership, and Men. S 222 236. Asch, S. E. (1955). Opinions and social pressure. Scientific American. 193(5), 31 35. Axsom, D., Yates, S., & Chaiken, S. (1987). Audience response as a heuristic cue in persuasion, Journal of Personality and Social Psychology 53, 30 40. Bandura, A. (1977). Social Learning Theory Engle wood Cliffs, NJ: Prentice Hall. Bandura, A. (2001). Social Cognitive Theory of mass communication. Media Psychology, 3 265 299. Baran, S. J., & Davis D.K. (2009). Mass communication theory: Foundations, ferment, and future (5th ed.). Boston: Wadsworth Pu blishing. Bernhardt, J. M., & Felter, E. M. (2004). Online pediatric information seeking among mothers of young children: Results from a qualitative study using focus groups. Journal of Medical Internet Research 6(1), e7. Retrieved June 18, 2007, from htt p://www.jmir.org/2004/1/e7 Beskow, L.M., Friedman, J.Y., Hardy, N.C., Lin, L., & Weinfurt, K.P. (2010). Simplifying informed consent for biorepositories: Stakeholder perspectives. Genetics in Medicine xx(xx), 1 6. doi:10.1097/gIM.0b013e318lead64d Bickart, B., & Schindler, R.M. (2001). Internet forums as influential sources of consumer information. Journal of Interactive Marketing 15(3), 31 40. Bickmore, T. W., Pfeifer, L. M., & Paasche Orlow, M. K. (2009). Using computer agents to explain medical document s to patients with low health literacy. Patient Education and Counseling 75(3), 315 320. Boninger, D.S., Krosnick, J.A., & Berent M.K. (1993). Origins of attitude importance: Self interest, social identification, and value relevance. Journal of Personal ity and Social Psychology 68(1), 61 80. Botan, C. (1997). Ethics in Strategic Communication Campaigns: The Case for a New Approach to Public Relations. The Journal of Business Communication, 34(2), 188 202.

PAGE 100

100 Bredenoord, A. L., Kroes, H. Y., Cuppen, E., Pa rker, M., & van Delden, J. J. (2011). Disclosure of individual genetic data to research participants: the debate reconsidered. Trends in Genetics 27(2), 41 47. Brenner, L.H., Brenner, A.T., & Horowitz, D. (2009). Beyond Informed Consent: Educating the Pat ient. Clinical Orthopaedics and Related Research 467(2), 348 351. Brock, T. C., & Green, M. C. (2005). Persuasion: Psychological Insights and Perspectives Thousand Oakes, CA: Sage. Burger, J.M., Soroka S., Gonzago, K., Murphy, E., & Somervell, E. (2001). The effect of fleeting attraction on compliance to requests. Personality and Social Psychology Bulletin .27, 1578 1586. Burns, R., & Burns, R. P. (2008). Business research methods and statistics using SPSS. Sage Publications Limited. Cacioppo J.T., & Petty R.E. (1982). The need for cognition. Journal of Persuasion and Social Psychology, 42, 116 131. Cacioppo, J. T., Petty, R. E., & Kao, C. F. (1984). The efficient assessment of need for cognition. Journal of Personality Assessment 48, 306 307. Cacioppo, J T., Petty, R. E., & Morris, K. J. (1983). Effects of need for cognition on message evaluation, recall, and persuasion. Journal of Personality and Social Psychology 45 (4), 805 818. Cacioppo, J. T., Petty, R. E., & Stoltenberg, C. D. (1985). Processes of social influence: The elaboration likelihood model of persuasion. Advances in cognitive behavioral research and therapy 4 215 274. Cacioppo, J. T., Petty, R. E., Feinstein, J., & Jarvis, W. B. (1996). Individual differences in cognitive motivation: The l ife and times of people varying in need for cognition. Psychological Bulletin 119, 197 253. Cacioppo, J.T., Petty, R.E., & Morris, K.J. (1983). Effects of n eed for cognition on message evaluation, recall, and persuasion. Journal of Personality and Social Psychology 45(4), 805 818. Cacioppo, J.T., Petty, R.E., Feinstein, J.A., & Jarvis, W.B.G.(1996). Dispositional differences in cognitive motivation: The life and times of individuals varying in need for cognition. Psychological Bulletin 119(2), 197 253. d oi: 10.1037/0033 2909.119.2.197 Cairncross, S., & Mannion, M. (2001). Interactive multimedia and learning: Realizing the benefits. Innovations in Education and Teaching International, 38(2), 156 164.

PAGE 101

101 Chaiken, S. & Maheswaran, D. (1994). Heuristic processin g can bias systematic processing: Effects of source credibility, argument ambiguity, and task importance on attitude judgment. Journal of Personality and Social Psychology 66(3), 460 473. Chaiken, S. (1979). Communicator physical attractiveness and persua sion. Journal of Personality and Social Psychology 37, 1387 1397. Chevalier, Judith A., & Mayzlin, D. (2006). The effect of word of mouth on sales: Online book reviews. Journal of Marketing Research 43 (August), 345 54. Choi, S.M., & Salmon, C.T. (2003). The elaboration likelihood model of persuasion after two decades: A review of criticism and contributions. The Kentucky Journal of Communication 22(1), 47 77. Cialdini, R. B. (1984). Influence: The Psychology of Persuasion New York: William Morrow. Cial dini, R.B., & Goldstein N.J. (2004). Social influence: Compliance and conformity. Annual Review of Psychology, 55, 591 621. Cohen, A.R., Stotland, E., & Wolfe, D.M. (1955). An experimental investigation of need for cognition. The Journal of Abnormal and S ocial Psychology 51(2), 291 294. Cruz, M. G., Boster, F. J., & Rodriguez, J. I. (1997). The impact of group size and proportion of shared information on the exchange and integration of information in groups. Communication Research 24, 291 313. De Vries, P., & Pruyn, A. (2007). Source salience and the persuasiveness of peer recommendations: The mediating role of social trust. In Persuasive Technology (pp. 164 175). Springer Berlin Heidelberg. Degner L.F., & Sloan J.A. (1992). Decision making during serious illness: What role do patients really want to play? Journal of Clinical Epidemiology 45, 941 950. Dellarocas, C. (2003). The digitization of word of mouth: Promise and challenges of online feedback mechanisms. Management Science 49(10), 1401 1424. Denn ett, D.C. (1987). The intentional stance. The MIT press, Cambridge, MA. Dolnicar, S., & Grn, B. (2007). How constrained a response: A comparison of binary, ordinal and metric answer formats. Journal of Retailing and Consumer Services 14(2), 108 122. Eas tin, M. S. (2001). Credibility assessments of online health information: The effects of source expertise and knowledge of content. Journal of Computer Mediated Communication 6.

PAGE 102

102 sire for autonomy Decision making and information seeking preferences among medical patients. Journal of General Internal Medicine 4, 23 30. Evans, L.M. & Petty, R.E. (2003). Self guide framing and persuasion: Responsibly increasing message processing to ideal levels. Personality and Social Psychology Bulletin 29, 313 324. Eysenbach, G., & Kohler, C. (2002). How do consumers search for and appraise health information on the World Wide Web? Qualitative study using focus groups, usability tests, and in dept h interviews. British Medical Journal 324(7337), 573 577. doi:10.1136/bmj.324.7337.573 Faden, R. R., & Beauchamp, T. L. (1986). A history and theory of informed consent. USA: Oxford University Press. Felt, U., Bister, M.D., Strassnig, M., & Wagner, U. (20 09). Refusing the information paradigm: Informed consent, medical research, and patient participation. Health 13, 87 106. Festinger, L. (1954). A theory of social comparison processes. Human Relations 7, 117 140. Fisher, R. J., & Ackerman, D. (1998). The effects of recognition and group need on volunteerism: A social norm perspective. Journal of Consumer Research 25 (3), 262 275. Flory, J., & Emanuel, E. (2004).Interventions to improve research participants' understanding in informed consent for researc h. The Journal of American Medical Association 292, 1593 1601. Fortun, P., West, J., Chalkley L., Shonde, A., & Hawkey, C. (2008). Recall of informed consent information by healthy volunteers in clinical trials. QJM, 101 625 629. doi:10.1093/qjmed/hcn0 67. Fox, S., & Jones, S. The social life of health information. Pew Internet 11 June 2009. Available from: www.pewinternet.org/Reports/2009/8 The Social Life of Health Information.aspx Frost, J. H., & Massagli M.P. (2008). Social uses of personal health i nformation within PatientsLikeMe, an online patient community: What can happen when patients have access to one another's data. Journal of Medical Internet Research 12(3), 21. doi:10.2196/jmir.1548 Gigone, D., & Hastie, R. (1993). The common knowledge eff ect: Information sampling and group judgment. Journal of Personality and Social Psychology 65, 959 974.

PAGE 103

103 Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2010). Multivariate data analysis (Vol. 7). Upper Saddle River, NJ: Prentice Hall. Haugtvedt, C. P., Petty, R. E., & Cacioppo, J. T. (1992). Need for cognition and advertising: Understanding the role of personality variables in consumer behavior. Journal of Consumer Psychology 1 (3), 239 260. Henningsen, M. L. M., Henningsen, D. D ., Cruz, M. G., & Morrill, J. (2003). Social influence in groups: A comparative application of relational framing theory and the elaboration likelihood model of persuasion. Communication Monographs 70 (3), 175 197. Houts, P. S., Witmer, J. T., Egeth, H. E. Loscalzo, M. J., & Zabora, J. R. (2001). Using pictographs to enhance recall of spoken medical instructions II. Patient Education and Counseling, 43(3), 231 242. Hu, Y., & Sundar, S. (2010). Effects of online health sources on credibility and behavioral intentions. Communication Research, 37 (1), 105 132. doi:10.1177/0093650209351512. Isenberg, D. J. (1986). Group polarization: A critical review and meta analysis. Journal of Personality and Social Psychology 50, 1141 1151. Jackson, C. L., Bolen, S., Branc ati, F. L., Batts Turner, M. L., & Gary, T. L. (2006). A systematic review of interactive computer assisted technology in diabetes care. Journal of General Internal Medicine 21(2), 105 110. doi:10.1111/j.1525 1497.2005.00310.x Janis, I. L. (1972). Victims of groupthink. Boston : Houghton Joffe, S., Cook, E.F., Cleary, P.D., Clark, J.W., & Weeks, J.C. (2001). Quality of informed consent: A new measure of understanding among research subjects. Journal of the National Cancer Institute 93, 139 147. Jsang, A., Ismail, R., & Boyd, C. (2007). A survey of trust and reputation systems for online service provision. Decision Support Systems 43, 618 644. Kalick, S.M. (1988). Physical attractiveness as a status cue. Journal of Experimental Social Psychology 24, 469 489. Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges and opportunities of Social Media. Business Horizons 53, 59 68. Kass, N., Chaisson, L., Taylor, H., Lohse, J. (2011). Length and Complexity of US and International HIV Consent Forms fro m Federal HIV Network Trials. Journal of General Internal Medicine 26(7).

PAGE 104

104 Kernisan, L.P., Sudore, R.L., & Knight, S.J. (2010). Information seeking at a caregiving website: A qualitative analysis. Journal of Medical Internet Research 12(3), 21. doi:10.219 6/jmir.1548. Kirsch, I., Jungeblut, A., Jenkins, L., & Kolstat, A. (2003). Adult Literacy in America: A First Look at the Findings of the National Adult Literacy Survey 3rd ed. Washington, DC: National Center for Education, U.S. Department of Education. L atan, B. (1981). The psychology of social impact. American Psychologist 36, 343 356. Levine, John M., Lauren B. Resnick, and E. Tory Higgins. "Social foundations of cognition." Annual Review of Psychology 44.1 (1993): 585 612. Levinson W., Kao A., Kuby A. & Thisted R.A. (2005). Not all patients want to participate in decision making. Journal of General Internal Medicine 20, 531 535. Lewis, D. (1999). Computer based approaches to patient education: A review of the literature. Journal of the American Med ical Informatics Association 6(4), 272 282. Liberman, A., & Chaiken, S. (1996). The direct effect of personal relevance on attitudes. Personality and Social Psychology Bulletin 22 (3), 269 279. Lohmann, S., Ziegler, J., & Tetzlaff, L. (2009). Comparison o f tag cloud layouts: Task related performance and visual exploration. Interact 1 (5726), 392 404. MacKie, Diane M., M. Cecilia Gastardo Conaco, and John J. Skelly. "Knowledge of the advocated position and the processing of in group and out group persuasiv e messages." Personality and Social Psychology Bulletin 18.2 (1992): 145 151. On Taking the Hippocratic Oath. New England Journal of Medicine 350, 2026 2029. McCroskey, J. C., Richmond, V. P., & Daly, J. A. (1975). The development of a measure of perceived homophily in interpersonal communication. Human Communication Research 1, 323 332. Miller, R., & Sim, I. (2004). Physicians' use of electronic medical records: Barriers and solutions. Health Affairs 23(2), 116 1 26. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research (1979). Belmont Report: Ethical principles and guidelines for the protection of human subjects of research Washington (D. C.): National Commission for the P rotection of Human Subjects of Biomedical and Behavioral Research.

PAGE 105

105 National Institute of Health (2011). Informed consent. In Glossary of NIH Terms Retrieved May 26, 2011 from http://grants.nih.gov/grants/glossary.htm#I Nelson Marten, P., & Rich, B. A. (19 99). A historical perspective of informed consent in clinical practice and research. Seminars in Oncology Nursing, 15 (2), 81 88. doi:10.1016/S0749 2081(99)80065 5. Paasche Orlow, M. K., Taylor, H. A., & Brancati, F. L. (2003). Readability standards for inf ormed consent: Forms as compared with actual readability. The New England Journal of Medicine, 348 721 726. Paris, A., Chaves, D.N., Cornu, C., Maison, P., Salvat Melis, M., Ribuot, C., Brandt, C., Bosson, J., Hommel, M., & Cracowski, J. (2007). Improveme nt of the comprehension of written information given to healthy volunteers in biomedical research: a single blind randomized controlled study, Fundamental and Clinical Pharmacology 21, 207 214. Park, D.H., Lee, J., & Han, I. (2007). The effects of on line consumer reviews on consumer purchasing intention: the moderating role of involvement, International Journal of Electronic Commerce 11 (4), 125 148. Write Fake Positive Amazon Reviews. The Daily Background Retrieved from http://www.thedailybackground.com/2009/01/16/exclusive belkins development rep is hiring people to write fake positive amazon reviews Peabody, D. (1962). Two components in bipolar scales: Direction and extremeness. Psychological Review 69(2), 65. Percy, L. (1976). An argument in support of ordinary factor analysis of dichotomous variables. Advances in consumer research 3(1), 143 148. Perloff (2010). The Dynamics of Persuasion: Communication and attitud es in the 21st Century (4th ed.). New York, NY: Routledge. Petty, R. E., & Cacioppo, J. T. (1981). Attitudes and persuasion: Classic and contemporary approaches Dubuque, IA: William C. Brown. Petty, R. E., Harkins, S. G., & Williams, K. D. (1980). The eff ects of group diffusion of cognitive effort on attitudes: An information processing view. Journal of Personality and Social Psychology 38, 81 92. Petty, R.E., & Cacioppo, J.T. (1986). The elaboration likelihood model of persuasion. In L. Berkowitz. Advanc es in experimental social psychology (123 205). New York: Academic Press.

PAGE 106

106 Petty, R.E., Cacioppo, J.T. & Goldman, R. (1981). Personal involvement as a determinant of argument based persuasion. Journal of Personality and Social Psychology 41, 847 855. Petty R.E., Wegener, D.T., & Fabrigar, L.R. (1997). Attitudes and attitude change. Annual Review of Psychology, 48, 609 647. Petty, R.E., Wegener, D.T., Fabrigar, L.R., Priester, J.R., & Cacioppo, J.T. (1993). Conceptual and methodological issues in the elabor ation likelihood model of persuasion: A reply to the Michigan State critics. Communication Theory 3(4), 336 342. Piolat, A., Olive, T., & Kellogg, R. T. (2004). Cognitive effort during note taking. Applied Cognitive Psychology 19 (3), 291 312. Prohaska, T R., & Glasser, M. (1996). Patients' views of family involvement in medical care decisions and encounters. Research on Aging 18 (1), 52 69. Ridpath,J.R., Greene, S.M., & Wiese, C.J. (2007). PRISM Readability Toolkit Seattle: Group Health Research Institu te. Available http://www.grouphealthresearch.org/capab ilities/readability/readability_home.htm l. Accessed 2009. Ries, N. M., LeGrandeur, J., & Caulfield, T. (2010). Handling ethical, legal and social issues in birth cohort studies involving genetic resea rch: Responses from studies in six countries. BMC Medical Ethics 11(1), 4. Rogers, E. M. (2003). Diffusion of Innovations New York: Free press. Ryan, R.M., & Deci, E.L. (2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology 25(1), 54 67. Seltman, H.J. (2009). Experimental Design for Behavioral and Social Sciences. Carnegie Melon University. Retrieved on February 20, 2011, from http://www.stat.cmu.edu/~hseltman/309/Book/Book.pdf Skloot, R. (2010). The Immortal Life of Henrietta Lacks New York: Random House. Smith, S.M., & Levin, I.P. (1996). Need for cognition and choice framing effects. Journal of Behavioral Decision Making, 9, 283 290. Soll, J. B., & Larrick, R. P. (2009). Strategies for revising judgment: How (and how well) people use others' opinions. Journal of Experimental Psychology. Learning, Memory, and Cognition, 35(3), 780. Steginga, S.K., & Occhipinti, S. (2004). The application of the heuristic systematic processing model to tre atment decision making about prostate cancer. Medical Decision Making 24, 573 589.

PAGE 107

107 Sugarman, J., McCrory, D. C., & Hubal, R. C. (1998). Getting meaningful informed consent from older adults: A structured literature review of empirical research. Journal of the American Geriatrics Society, 46 (4), 517 524. Sundar, S.S., & Nass, C. (2001). Conceptualizing sources in online news. Journal of Communication 51(1), 52 72. doi: 10.1093/joc/51.1.52. Suzuki, L. K., & Calzo, J. P. (2004). The search for peer advice in cyberspace: An examination of online teen bulletin boards about health and sexuality. Applied Development Psychology 25, 685 698. Tanis, M., & Postmes, T. (2003). Social cues and impression formation in CMC. Journal of Communication 53 (4), 676 693. Trum bo, C.W. (1999). Heuristic systematic information processing and risk judgment. Risk Analysis 19, 391 400. United Nations (1948). Universal declaration of human rights. Resolution adopted by the General Assembly, 10(12). Vollmann, J., & Winau, R. (1996). Informed consent in human experimentation before the Nuremberg code. British Medical Journal 313, 1445 1449. Walther, J.B, Zuoming, W., & Tracy, L. (2004). The effect of top level domains and advertisements on health web site credibility. Journal of Medi cal Internet Research 6(3):e24. Walther, J.B., DeAndrea, D.,Kim, J., & Anthony, J.C. (2010). The influence of online comments on perceptions of antimarijuana public service announcements on YouTube. Human Communication Research 36(4), 469 492. Wang, Z., Walther, J. B., Pingree, S., & Hawkins, R. (2008). Health information, credibility, homophily, and influence via the Internet: Web sites versus discussion groups. Health Communication 23, 358 368. Wegener, D. T., Downing, J., Krosnick, J. A., & Petty, R. E. (1995). Measures and manipulations of strength related properties of attitudes: Current practice and future directions In R. E. Petty & J. A. Krosnick (Eds.), Attitude strength: Antecedents and consequences (455 487). Mahwah, NJ: Lawrence Erlbaum Assoc iates. Williams Piehota, P., Pizarro, J., & Schneider, T.R. (2003). Matching health messages to information processing styles: Need for cognition and mammography utilization. Health Communication, 15 (4), 375 392. Wittenbaum, G. M., Hubbell, A. P., & Zucke rman, C. (1999). Mutual enhancement: Toward an understanding of the collective preference for shared information. Journal of Personality and Social Psychology 77, 967 978.

PAGE 108

108 Wu, C. & Shaffer, D.R. (1987). Susceptibility to persuasive appeals as a function o f source credibility and prior experience with the attitude object. Journal of Personality and Social Psychology 52(4), 677 688. Zelinski, E.M., Crimmins, E., Reynolds, S., & Seeman, T. (1998). Do medical conditions affect cognition in older adults? Heal th Psychology 17(6), 504 512. Conference on Information Systems (ICIS). Milwaukee: Association for Information Systems. Zhang, Y. (1996). Responses to humorous advertising: The moderating effect of need for cognition. Journal of Advertising 15 32.

PAGE 109

109 BIOGRAPHICAL SKETCH Yukari Takata Schneider was born in Yamaguchi, Japan Her parents are ethnolingu ists and have been involved with international missions all her life. As a result, she spent much of her informative years in rural parts of Indonesia and the Philippines and was blessed with friendships with people of diverse backgrounds and worldviews. Through her travels, she also gained a hunger for learning more about this beautiful world. She attended Covenant College in Lookout Mountain, Georgia and graduated with a Bachelor of Arts in b iology in 2005. For two years she worked at the Tennessee Aquar ium where she was responsible for the seahorses. Every day was filled with fascinating moments but she was drawn to learning more about people, so she returned to school and received her Master in Public Health in 2007 from the University of Florida Colleg e of Public Health and Health Professions. spent some time in Kenya exploring its natural wonders, but feeling the hunger to learn more, she returned to the U niversity of Florid a to pursue a Doctor of Philosophy in m ass c ommunication in 2009 Since then, she has explored how social forces of influence are expressed through mass media, especially as it relates to health. And in 2010, she received a dissertation fellowship from the Social Science Research Council to support her studies. In 2012 she relocated to Baltimore, Maryland after receiving a position with a health communications company, and married her champion and muse, Logan Schneider.