Cognition and Learning in Decision-Making as Compensatory Mechanism for Emotional Processing Deficit

Introduction

Experiments by Bechara et al. (1999) investigated the performances on the Iowa Gambling Task (IGT) of two groups of patients with either their ventromedial prefrontal cortex (VMPF) or amygdala damaged, predicting their outcomes to be based on distinct mechanisms of impaired somatic markers and related neural components and circuits required for good decision-making. The group of patients with damaged amygdala also exhibited low skin conductance response (SCR) because of a dysfunctional region, or lack thereof, which is responsible for emotion-guided processing influencing decision-making. The premise of this study has been challenged by a following experiment by Fellows and Farah (2005), which suggested the point of laterality in VMPF impairment, the questionable role of dorsolateral prefrontal cortex (DLPF) and reversal learning as factors contending the somatic marker hypothesis.

The conflicting evidence regarding DLPF's role in decision-making with its damage contributing to worst performance in IGT in this experiment was not previously tested (Bechara et al., 1999). In contrast, a shuffled variant of IGT developed by Fellows and Farah (2005) examined the role of reversal learning, a form of flexible stimulus-reinforcement learning, in performances by two groups of patients with either VMPF or DLPF damaged. In this shuffled variant, the cards with losses, be they large or small, were shuffled and placed first in position on all decks of cards with either high risk and magnitude of losses (disadvantageous) or low risk and magnitude of losses (advantageous).As a result, their hypothesis was confirmed by VMPF patients performing well in shuffled variant version, whereas DLPF patients did not.

Discussion

There are notable differences in key findings and conclusions between the two aforementioned studies. Bechara et al. (1999) demonstrated that VMPF and amygdala patients performed worse than controls in the IGT, whereas SCR was preserved in VMPF patients and controls but not amygdala patients. In contrast, Fellows and Farah (2005) noted that left hemisphere predominant damage VMPF patients chose more cards from risky disadvantageous decks. Along with DLPF patients, they performed worse on IGT than controls but only DLPF patients did worst in the shuffled variant form of IGT. VMPF patients benefited from the shuffled variant but still performed poorly in a second IGT, implying that they are still capable of advantageous decision-making despite impaired somatic markers and VMPF inactivity.

In terms of challenging the somatic marker hypothesis, which was supported in the experiments by Bechara et al. (1999), the study results by Fellows and Farah (2005) suggested that cognitive deficit in the form of reversal learning, which could have negatively affected good decision-making, was not previously considered as a potential contributing factor in the somatic marker hypothesis. This “contingency reversal” learning was later clarified to be a component that requires somatic marker signals to work (Bechara et al., 2005). However, in emphasizing the discrete and independent role of somatic aroused emotion in ensuring efficient choice and decision-making, the contributors of thought and cognition and part of the functional processes activated in the brain were precluded and ignored.

Although a region-specific attribution of reversal learning deficit points toward the VMPF and numerous studies have supported its key function in moral and emotional judgment and decision-making, as well as empathy in social behaviour, it may undermine concurrent involvement of other brain regions in related functions that share close connections with the VMPF, for example, hippocampal region, as later studies have found comparable impaired performance in the IGT by hippocampal damaged patients (Ciaramelli et al., 2007; Gupta et al., 2009; Gutbrod et al., 2006; Shamay-Tsoory et al., 2003). The disproportionate appreciation of emotional marker signals in somatic marker hypothesis could unintentionally overlook and undervalue the significance of memory and cognition contribution in judgment and decision-making, which could be better assessed by the task performance of patients having both VMPF and hippocampal lesions instead of either one of these two regions separately, as reported by McCormick et al. (2018).

Although the generation of emotional somatic marker signals may be timely on cue and loss-saving, an emotionally arousing signal may not come laid out with information content that has sufficient clarity and specificity. This implies that cognition and learning add to a more integrative model of the decision-making process and that somatic markers are not the sole players involved to deter us from making risky high-cost decisions. Since there was a choice between large and small immediate losses in the decks of cards under both the IGT and its shuffled variant, the possibility of an implicit processing of anticipation for future larger gains could be indirectly occurring. Such implicit cognitive processing that was not explicitly investigated involves discounting immediate losses for future potentially larger gains and such immediacy discounting is likely not impaired by VMPF dysfunction.

Nevertheless, Bechara et al. (2005), in addressing a critique study by Maia and McClelland (2005) on somatic marker hypothesis, contended that emotion-related marker signals invoked by internal neurophysiological processes, of which the VMPF and amygdala are involved, function as a necessary initiator and assistant to optimal choice and decision-making that maximizes gains and minimizes losses. Intuitively, the direction that such emotional marker signals is taking the individual helps maintain a personally relevant focus because the signals generated internally are intimate, self-assuring, and a direct first-hand experience that is as proximal as the “gut.” This could explain why anticipatory SCR is very easily maintained during the IGT experiments for healthy participants.

On the other hand, it also questions the lead up by anticipatory SCR to potential surplus of emotional arousal and predictably fear that could be over-reactive and disproportionately experienced by different genders (refer Tranel et al., 2005), which should we consider as distinct from the processing of conscious knowledge of the situation during the conceptual period (as stated by Bechara et al., 2005), may produce counterproductive outcomes. In the event that such might happen, would an overactive role played by emotional marker signals that is unguided by discrete “rational” conscious cognitive processing of the situation in real time produce equally beneficial choice and decision-making outcomes?

In contrast, would it be more justified to say that the function of emotional somatic markers is at best, complementary, mutual, and assistive to personally rewarding choice and decision-making that work nondisruptive in concert and humble cooperation with conscious cognitive and knowledge processing?. Bechara et al. (2005) has claimed that there is a break and mutual exclusivity between nonconscious somatic signal processing and conscious knowledge of the situation based on the findings of anticipatory SCR. However, anticipatory SCR is only one of a number of measurable physiological outcomes expressed by emotional marker signaling.

We need more robust assessments of actual neurophysiological initiation and processing, as opposed to solely outcome, that is, real-time functional magnetic resonance imaging (fMRI) focused on VMPF and amygdala regions of interest, to be certain that the highly compact and interconnected human brain is capable of temporally discrete nonconscious somatic signaling and conscious specific information and knowledge-oriented cognitive processing, as claimed by the somatic marker hypothesis.

Conclusion

In the experiments by Bechara et al. (1999), the comparable performances of VMPF and amygdala patients imply that the impairment produced by damage to either one of these two regions can be compensated by the other, which is inadequately explained by the hypothesis that proposed VMPF and amygdala were responsible for different mechanisms in emotional decision-making with both groups' performances overlapping at just about the same level. It would be reasonable to not preclude the role of fear of future losses in the IGT as a heuristic judgment for decision-making in patients with intact and normal functioning amygdala region. Inasmuch as the amygdala is involved for VMPF patients, the chances of fear conditioning being activated, although not explicitly stated and tested in this study's paradigm, should be above random in increasing the frequency of low-risk choices from the advantageous card decks.