Cognitive sources of evidence for neuroticism's link to punishment-reactivity processes

Abstract

Theories of neuroticism emphasise its close potential link to punishment-reactivity processes, yet cognitive sources of evidence for this proposed processing basis are surprisingly scarce. The present two studies (N = 123) sought to rectify this important gap in the literature in terms of reactivity to error feedback. Study 1 found that individuals high in neuroticism were faster to switch behavioural responses following errors, whereas an opposite pattern was found among individuals low in neuroticism. Study 2 extended this error-reactivity perspective to the realm of behavioural decision making. Individuals high in neuroticism switched their behavioural predictions following error feedback, whereas this tendency was non-significant among individuals low in neuroticism. Together, the studies present novel, but theory-informed, cognitive paradigms for assessing punishment-reactivity processes, confirm neuroticism's link to such processes, and do so in the realms of both reaction time (Study 1) and behavioural predictions (Study 2). The discussion focuses on the utility of modelling punishment-reactivity processes in cognitive terms and highlights relevant directions for future research.

Keywords:

• Neuroticism
• Punishment
• Reactivity
• Affect
• Cognition
• Decision making

Notes

¹It was neither adaptive nor problematic to switch behavioural responding following error feedback in the present paradigms, due to the randomised nature of trials. Further, we regard it unlikely that strategic factors played a role in the paradigms in that both used “continuous priming” procedures shown to effectively guard against expectancy effects (McRae & Boisvert, 1998; Robinson, Moeller, & Goetz, 2009; Shelton & Martin, 1992). That is, from the participant's perspective, each trial was an independent one requiring an independent response. Error-reactivity effects would therefore reflect automatic processing biases, hypothesised to differ by levels of neuroticism, rather than strategic considerations or expectancy effects.

²Within-subject designs of a cognitive type are fairly powerful and even smaller samples are often typical to the cognitive literature (Sanders, 1998). From a personality-processing perspective, however, the sample size was smaller than desired. This can be attributed to the relative difficulty that we had obtaining study volunteers at the time and the procedures, in which participants were run individually. A larger sample size was collected in Study 2 and thus we will particularly emphasise results that conceptually replicate across the two studies.

³Following the cognitive literature (Sanders, 1998), and extensions of it to understanding personality processes (Robinson, 2007b), results should be interpreted in terms of the within-subject effect of the error feedback manipulation at a given level of punishment sensitivity, either low (−1 SD) or high (+ 1 SD). Doing so ensures that each participant serves as his or her own control and thus eliminates unwanted sources of variance—such as baseline tendencies to switch responding across trials, irrespective of error feedback—that are likely to be confounded with the personality trait of interest in a somewhat random way from sample to sample. In other words, further development of processing theories of personality should capitalise on the logic and benefits of the within-subject design, which is fundamental to cognitive science (Canli, 2004).