The effect of cognitive and emotion-based processes on the Iowa Gambling Task

Real life decision-making depends on a complex interplay between cognitive and emotion-based processes. Damasio (1994) developed the Somatic Marker Hypothesis (SMH) arguing that emotion-based processes guide decision-making by directing individuals towards alternatives that have been previously ‘marked’ as positive or guide them away from the negative options. The primarily used test-bed of the emotion-based learning is Iowa Gambling Task (IGT, Bechara, Damasio, Damasio, & Anderson, 1994). The SMH makes three assumptions about the IGT behaviour: (a) somatic markers have a negative connotation and bias decision-making covertly in the absence of explicit knowledge, (b) there is a limited role for cognitive procesesing during IGT performance, especially during the initial stages of the task, and (c) anticipatory somatic markers guide decision-choices away from the bad options as participants are able to anticipate the good and the bad options. This thesis tested the SMH using a combination of psychophysiological methods (Eye-tracking, Pupillometry, Heart Rate and Blood Pressure measurements), behavioural measurements and psychometric measures of individual differences in combination with the IGT. The systematic review, meta-analyses and the experiments described in this Thesis explored the validity of these assumptions and found that they are not accurately manifested in behaviour during IGT performance. A novel methodology not previously employed was used to capture somatic markers through pupillary responses. Explicit learning was also assessed by the eye-tracking methodology in testing IGT performance in normal conditions and under stress. The results from the first two experiments indicated that explicit processing and knowledge about the task are more critical factors during the early stages of the game than previously suggested. Although there were some indicators of the existence of somatic markers, it was found that cognitive reflection, conscious awareness and increased cognitive processing occurred early in the game and guided behaviour on IGT. The results from the final experiment revealed that IGT performance in healthy individuals is not always optimal; stress levels impaired performance whereby a lack of, or insufficient cognitive processing early in the game may create a somatic signal that interferes with IGT performance. Furthermore, attentional processing, cognitive reflection and conscious awareness can be disrupted by stress resulting in non-optimal decision-making strategies that consequently interfere with performance on the IGT. Taken together, these results challenge the basic premises of the SMH and could be best explained within the dual-process framework (e.g., Brevers, Bechara, Cleeremans, & Noel, 2013). If somatic markers do not play a significant role in learning IGT than the task needs to be re-evaluated and caution is warranted when the IGT is used as a diagnostic tool to measure decision-making deficits in clinical populations.