During the past three decades, self-efficacy (SE) has emerged as a widely examined psychological influences on chronic pain and related functional outcomes.Bandura [13-15] defined SE as confidence that one can successfully execute a course of action to produce a desired outcome in a given situation and contended SE determines how much effort and persistence people exhibit in the face of obstacles or aversive experiences.Although numerous studies have linked SE with more adaptive functioning in chronic pain samples, some studies have reported null effects of SE on functioning.Given inconsistencies in the research on SE and adjustment in chronic pain samples, the main purpose of this research was to perform a meta-analysis to (1) evaluate overall strengths of relation between SE and key outcomes of chronic pain (i.e., pain severity, functional impairment, affective distress) and (2) to identify potential moderating effects of specific sample characteristics and methodologicai factors on these associations.A systematic search of the literature on SE and chronic pain resulted in the initial consideration of 1688 studies.Screening of these studies for appropriate papers resulted in the identification of 83 studies comprising 86 samples (N=15,616) that fulfilled selection criteria for inclusion in the meta-analysis.Data analyses indicated SE had significant negative overall correlations with impairment (r=-0.49, p＜0.001, 95％ CI:-0.55 to-0.43), affective distress (r=-0.43, p＜0.001, 95％CI:-0.47 to-0.39), and pain severity (r=-0.39, p＜0.001, 95％ CI:-0.44 to-0.34).However, high levels of heterogeneity were evident in analyses of overall relations between SE and each outcome.Moderator analyses indicated age, pain duration, SE scale content (i.e., SE for functioning despite pain versus SE for pain control versus SE for managing other symptoms such as emotional distress) and type of impairment measure (self-report versus performance) had significant moderating effects on SE-impairment associations.SE-affective distress relations were also moderated by employment status and SE scale content.However, moderator analyses failed to identify sample or methodological factors that helped to explain variability in effect sizes reflecting SE-pain severity relations.Notably, moderator analyses of studies having longitudinal designs indicated overall strengths of relation between SE and each outcome at follow-up remained significant within subsets of studies that had statistically controlled for effects of baseline responses on that outcome.As such, analyses indicated SE is both a robust correlate of key outcomes related to chronic pain and a potentially important risk/protective factor that has implications for subsequent functioning in affected groups.Finally, results of trim and fill analyses conducted to adjust for effects of publication bias indicated that adjusted effect sizes of SE and each of the three outcomes were very similar to unadjusted effect sizes.A secondary purpose of this research was to address the lack of experimental research designed to assess the causal impact of SE on pain perception.In Study 2, samples of 84 undergraduate women were randomly assessed through experimental manipulations in which they were given false feedback about their performance on a baseline cold pressor test.Participants randomly assigned to the high performance condition (n=27) were told "Just before you do the second cold pressor test, Ill give you a little feedback about how well you performed on the first cold pressor test.I wont tell you the exact amount of time you lasted but I can say that you lasted for a longer time than 91％ of people who have done the cold pressor test at SWU.You seem to be more capable than most people at the rusk" while those randomly assigned to the low performance condition (n=30) were told "Just before you do the second cold pressor test, Ill give you a little feedback about how well you performed on the first cold pressor test.I wont tell you the exact amount of time you lasted but I can say that you lasted for less time than 91％ of people who have done the cold pressor test at SWU.You seem to be less capable than most people at the task".Women randomly assigned to the average performance control condition (n=27) were told "Just before you do the second cold pressor test, Ill give you a little feedback about how well you performed on the first cold pressor test.I wont tell you the exact amount of time you lasted but I can say that you lasted for as much time as exactly half of people who have done the cold pressor test at SWU.You seem to have average capability compared to other people at the task".It was hypothesized that women assigned to the high performance feedback condition would report an increase in SE and show a significant improvement in pain tolerance on a follow-up CPT compared to women in the control condition.Furthermore, women assigned to the low performance feedback condition were expected to report a decrease in SE yet show a significant improvement in pain tolerance on a follow-up CPT compared to controls.Repeated measure ANOVA showed that there was significant interaction effect between self-efficacy and condition, F (2, 81) =3.91, p=0.024.Specifically, SE changed at the different way among high SE feedback, low SE feedback and average SE feedback.Repeated measure ANOVA also showed that the overall sample rated pain intensity at the same level on the 2nd CPT compared to the 1st CPT, F(1, 81)=2.578, p=0.11.There is a marginal significant pain tolerance x Condition interaction, F (2, 81) =3.06, p=0.052.Basically, low SE feedback resulted in a larger increase in pain tolerance between baseline and post-feedback CPTs compared to average or high SE feedback.The results of this study did not provide demonstration of our predicted effect, SE feedback that reduces self-efficacy can boost performance during the CPT.Basically, in response to low SE feedback; participants lowered their self-efficacy and showed increased level of performance.