Hot Thought: Mechanisms and Applications of Emotional Cognition

Canadian Psychology, Feb 2008 by Anderson, Adam K

Hot Thought: Mechanisms and Applications of Emotional Cognition, by Paul Thagard. The MIT Press, 2006, 313 pages (ISBN: 978-0-262-20164-3, CA$36.00 Cloth)

Reviewed by ADAM K. ANDERSON

DOI : 10.1037/0708-5591.49.1.68

Since the beginnings of early western philosophy, human beings have been posited to contain two fundamentally independent faculties, those of "passion" and "reason." Arousing the passions was viewed as antecedent to and not a consequence of reason. The province of reason was to harness or control the unruly passions. This historical segregation of emotional and cognitive functions has found reasonable support from more recent psychological and neuroscientific inquiry. There is, however, an increasing appreciation that emotion and cognition are mutually intertwined and thus should not be conceived of or studied with complete independence. Hot Thought is a cognitive scientist's attempt to bridge the narrowing gap between the study of thinking and feeling by examining and modelling emotional cognition-the study of how emotion and cognition interact to shape human thinking.

It is important to distinguish this work from a topical book on the emotions. As it gives short treatment of the theoretical and empirical work on the emotions, it would leave a student of the emotions wanting. This is not to say the book misses its mark, but rather its mark lies elsewhere. It is thus appropriate for both upper-level undergraduates and graduate students, particularly in cognitive science, who would benefit from a consideration of the role of the emotions and how to integrate them into models of thinking and reasoning. The role of emotions is largely confined to considering how inputs from motivated systems-systems that give rise to emotional valence (i.e., the positive and negative)regulate decision making. Building on the author's prior models of coherence of mental representations in decision making, the current work tackles how choices are not only made on the basis of maximal cold coherence between mental propositions, but also now considers one's emotional dispositions toward mental propositions, to maximise coherence in reasoning. It is in this sense that the author speaks of "hot cognition."

The book presents a collection of articles on the topic of emotional cognition written by the author since the turn of the century. The first half is devoted to the development of the neurocognitive coherence model for understanding the mechanisms of emotioncognition interactions. Subsequent chapters consider the broad application of this model, from base tabloid material ("Why Wasn't O.J. Convicted?") to lofty Kantian inspired assessments of theoretical reasoning ("Critique of Emotional Reason"). As it is a collection of independently written papers, where a model is developed and applied in various domains, content is necessarily repeated across chapters. This redundancy may be off-putting to some; however, it allows the reader to become acquainted with the central concepts of the book from multiple vantage points, as well as their development over a 5-year period of published articles. For instance, revisions of models of emotion-cognition interactions are presented. This includes multiple versions of the hot Coherence (HOTCO) model from an old-style prepositional to a connectionist implementation to a neurobiologically rigorous incarnation in GAGE, in reference to the eponymous patient rendered with dysfunctional decision making following damage to the emotional sectors of the prefrontal cortices. Along the way the model is applied to various levels of analysis-from how biologically modeled neural entities that take into account the electrophsyiologic properties of cortico-limbic interactions forge emotional influences on an individual's choice to how emotions influence the dynamics of social networks in group decision making.

The effect of reading various incarnations of emotion-cognition interactions, reflecting the author's maturing vision, is to afford a rare glimpse into the cognitive scientist's studio, as a theory is sculpted from raw materials (i.e., HOTCO 1) into a fully fleshed out work (GAGE). This shaping of theory is both an intellectually and emotionally satisfying enterprise. It is in this spirit that the collection ranges from the more cold content of mental representations and model development to later more emotionally involving chapters devoted to how to become a successful scientist, the role of positive and negative emotions in scientific discovery in Nobel Prize-winning scientists, and the importance of emotional coherence to religious belief. If one fails to read through the table of contents before embarking on reading, the trajectory of the book is a great surprise. Given the range of intellectual territory, the book has something for everyone.

The literal and metaphorical heart of the book is the notion of coherence and how emotions influence computations of coherence. Coherence borrows from connectionist networks the notion of maximal satisfaction of multiple constraints. To the degree that elements cohere, they will be positively coactivated; to the degree they incohere, they will negatively deactivated. The balancing of these multiple decision-making scales is a computationally intensive problem, likely only the tip of which achieves consciousness. Modification of this coherence to accommodate "hot cognition" is to allow emotion to tip the decision-making scales, affording additional negative and positively valenced weights for the to-beavoided and to-be-desired events. In this scheme it is unclear whether one should best consider emotions as causes or consequents, or both. Emotions may begin as feedback signals originating from weighing decisions that maximise or minimise coherence. Emotions may also represent the biases in these coherence calculations from the get-go, and thus are the primary movers of spreading of activation among competing alternatives, being one of the causes of coherence itself. The latter view suggests the influence of colder "attitudes" that are the latent negative and positive valences of events. The former is more akin to hot somatic markers that embody the decision-making process as they bias the contents of awareness. Such a formulation would need to consider the growing literature on how the experience of positive and negative emotions not only reflects the acceptance or rejection of alternatives, but also differs in information processing styles. Negative emotions are thought to restrict the scope of considered alternatives and has been associated with a more analytic decision making style. By contrast, positive emotions are thought to increase cognitive flexibility by generating a broader more exploratory mindset, casting a wider cognitive net during decision making. These differences are not captured in the view that people simply approach things they like and avoid what they dislike. The reader should look forward to future implementations of emotional coherence models that may consider these critical qualitative affective components to rational choice.