Neural networks underlying the metacognitive uncertainty response

Abstract

Humans monitor states of uncertainty that can guide decision-making. These uncertain states are evident behaviorally when humans decline to make a categorization response. Such behavioral uncertainty responses have also defined the search for metacognition in animals. While a plethora of neuroimaging studies have focused on uncertainty, the brain systems supporting a volitional strategy shift under uncertainty have not been distinguished from those observed in making introspective post-hoc reports of categorization uncertainty. Using rapid event-related fMRI, we demonstrate that the neural activity patterns elicited by humans’ uncertainty responses are qualitatively different from those recruited by associative processes during categorization. Participants performed a one-dimensional perceptual-categorization task in which an uncertainty-response option let them decline to make a categorization response. Uncertainty responding activated a distributed network including prefrontal cortical (PFC), anterior and posterior cingulate cortex (ACC, PCC), anterior insula, and posterior parietal areas; importantly, these regions were distinct from those whose activity was modulated by task difficulty. Generally, our results can be characterized as a large-scale cognitive control network including recently evolved brain regions such as the anterior dorsolateral and medial PFC. A metacognitive theory would view the uncertainty response as a deliberate behavioral adjustment rather than just a learned middle category response, and predicts this pattern of results. These neuroimaging results bolster previous behavioral findings, which suggested that different cognitive processes underlie responses due to associative learning versus the declaration of uncertainty. We conclude that the uncertainty response represents an elemental behavioral index of metacognition.

Occipitotemporal Category Representations Are Sensitive to Abstract Category Boundaries Defined by Generalization Demands

“…Additionally, whereas A/B tasks tend to elicit activity in frontoparietal and hippocampal regions (Seger et al, 2000;Zeithamova et al, 2008), A/notA tasks tend to elicit activity in visual cortices and in the basal ganglia (Reber et al, 1998(Reber et al, , 2003Aizenstein et al, 2000;Summerfield and Koechlin, 2008;Zeithamova et al, 2008). Regions associated with A/notA categorization thus closely resemble those associated with perceptual priming and repetition suppression (Wiggs and Martin, 1998;Koutstaal et al, 2001;Henson, 2003), as well as the theorized neurobiological substrate of the perceptual representation system (Schacter, 1990;Reber and Squire, 1999;Ashby and O'Brien, 2005;Casale and Ashby, 2008).…”

“…To identify neighborhoods of voxels representing distance from the prototype or distance from the decision bound, we used a searchlight approach (sphere radius 10 mm) (Kriegeskorte et al, 2006), in conjunction with linear support vector regression (SVR). We implemented the searchlight using custom code based on the Nilearn python package (Abraham et al, 2014) and implemented the SVR analysis using the SciKit-Learn machine learning package for python (Pedregosa et al, 2011), setting the SVR penalty parameter, C, to 0.01 based on the results from a separate dataset.…”

“…Although these voxels were selected based on this effect, visual inspection of the plots suggested an interaction between rule and distortion level. We therefore conducted a post hoc analysis (Friston et al, 2006;Kriegeskorte et al, 2009), which was orthogonal to the contrast used for voxel selection. A repeated-measures ANOVA indicated that the interaction was significant for both the left (F (3,45) ϭ 9.48, p Ͻ 0.01, 2 ϭ 0.39) and right (F (3,45) ϭ 11.76, p Ͻ 0.01, 2 ϭ 0.44) fusiform gyri, suggesting that the visual characteristics of the stimuli were processed differently between categorization rules.…”

“…In typical instantiations of the A/notA task, participants learn to apply a single generalization threshold (but see Nosofsky et al, 2012), and so neural signals reflecting representational factors (Strange et al, 2005;Seger et al, 2011Seger et al, , 2015Davis et al, 2014), which vary with distance from the prototype, and decisional factors (Grinband et al, 2006;Kayser et al, 2010;White et al, 2012), which vary with distance from the category boundary, are confounded, in that decisional difficulty increases with distance from the category prototype. In the current experiment, participants categorized filled dot prototype stimuli as category members or nonmembers according to "strict" and "lax" generalization rules (see Fig.…”

Emergence in extinction of enhanced and persistent responding to ambiguous aversive cues is associated with high MAOA activity in the prelimbic cortex

“…The present findings suggest that a similar relationship may exist for the response to ambiguous threat cues in non-social situations. In relation to the aforementioned “conditioned cognitive generalization”, human neuroimaging studies have suggested that with uncertainty individuals may adopt a (meta)cognitive coping state, associated with prefrontal cortex activity (Paul et al., 2015). Additionally, a relationship between intolerance to uncertainty and poorer extinction of unambiguous fear cue associated with greater amygdala and vmPFC activation was previously reported in human work (Morriss et al., 2015).…”