Autism is a neurodevelopmental disorder featuring severe social and communicative deficits. Cognitive impenetrability is the inability of systems to be affected by an individual’s beliefs and desires. Individuals with autism have delayed performance on "mindreading" tasks. Mindreading is the interpretation of others which enables successful social interactions. This task performance is not correlated with "everyday" social function in autism. These deficits are in contrast with general intelligence and reasoning abilities within the normal range, resistant to cognitive behavioral training, and partially attributable to atypical social predispositions. This suggests individuals with autism perform social tasks differently than typicals. I argue that this mindreading deficit is cognitively impenetrable, then position this deficit within a theoretical framework, enactivism, emphasizing cognitive embodiment and inseparability of social comprehension and action. This framework helps explain the cognitive impenetrability of these social deficits by "locating" failure to penetrate in individuals' failure to properly interact with the environment.

In this thesis I argue that a leading sentimentalist theory, Rational Sentimentalism, faces the Problem of Superfluity because the evaluative properties to which certain emotions are responses can be defined independently of examining those emotional responses. In other words, the connection to value that Rational Sentimentalism aims for fails to obtain. I show that at least one such emotion, disgust, is influenced by higher cognition to a degree incompatible with Rational Sentimentalism avoiding the Problem of Superfluity. I conclude by suggesting ways in which other emotions are structurally similar to disgust, and therefore face the same problem in being incorporated into Rational Sentimentalism.

When trying to make sense of uncertain situations, we might rely on summary information from a description, or information gathered from our personal experience. There are two approaches that both attempt to explain how we make risky decisions using descriptive or experiential information—the cognitive-based explanation from the description-experience gap, and the emotion-based explanation from the somatic marker hypothesis (SMH). This dissertation brings together these two approaches to better understand how we make risky decisions. Four options were presented, with options differing in terms of advantageousness and riskiness. How easy or difficult it was to consciously comprehend the reward structure, or cognitive penetrability, was manipulated by displaying single outcomes or multiple, diverse outcomes per trial. Within the description or experience task, participants were randomly assigned to the more or less penetrable version of an all gain or all loss set of options. How often the riskier or advantageous options were chosen served as a measure of risky or advantageous decision making. Regardless of penetrability, risk preferences were generally but not completely as predicted by the SMH. Instead, the primary effect of cognitive penetrability was on advantageous decision making. Furthermore, description was found to be more cognitively penetrable than experience. Overall, the results suggest that clarification is needed regarding how somatic markers are formed in the loss versus gain domain, and future research should consider the difference in penetrability between description and experience when trying to explain preferences between the two decisions.

Mental images are unique mental representations and the depictive view in the imagery debate states that mental images have similar spatial structures as their corresponding external object. The propositional view, however, contends that beliefs about the external world influence a mental image – also known as the cognitive penetrability theory –, therefore mental images cannot be described in visual terms. People with self-described aphantasia, those considered lacking visual mental images, offer a new opportunity to approach this issue. The current study employed the mental rotation task (MRT), where a three-dimensional object needs to be mentally aligned with another one that is rotated to a different angular position. To test the effects of beliefs on mental images, different instruction conditions were used on the MRT. Twenty-seven participants (21 females, mean age 47 years) with self-described aphantasia completed an online experiment, consisting of the MRT, the Object-Spatial Imagery Questionnaire (OSIQ), and the Vividness of Visual Imagery Questionnaire (VVIQ). While participants scored low on the VVIQ and higher on the spatial items than on the object items of the OSIQ, the key finding was that increasing reaction time with increasing angular disparity on the MRT was not influenced by different instruction types. This suggests that the theory of cognitive penetrability is not applicable on mental rotation in aphantasia, highlighting the importance to revise how mental images (or the lack of them) are understood and described.