Despite the privileged access we enjoy to our stream of thought, we often find it difficult to explain why a certain thought came to mind. Understanding the factors affecting the organization and regulation of thought is important given the evident relationship between thoughts, behavior, and wellbeing. However, the covert nature of thought processes makes them challenging to measure without relying solely on subjective introspection, or on their filtered, verbal or behavioral expressions. For instance, many disorders (e.g., obsessive-compulsive disorder) are characterized by the experience of uncontrollable, unwanted thoughts. However, whether such subjective experiences indicate an objective impairment in thought control is unclear and requires a better understanding of how we normally control our thoughts. In other cases (i.e., formal thought disorder), incoherent speech is used as a proxy for thought disorganization. Still, current methods do not allow to dissociate impairments in the basic organization of thought, from abnormalities in the control processes regulating the expression of (possibly normal) loose associations. One approach for tackling these questions relies on the basic idea that some (but not all) control processes are time-costly. Building on this idea, I will present a novel, tractable modeling approach integrating reported thoughts (e.g., associations) and response times (RTs) to analyze the covert organization and control of thought. I will demonstrate how this semi-Markov process (SMP) model can help gain insight into two different transdiagnostic dimensions. First, I will present data concerning the involvement of proactive and reactive thought control processes in a modified free association task, wherein each cue is presented several times, and participants are asked not to repeat associations. The results indicate that although people primarily react to unwanted thoughts after they occur, people also utilize two proactive control mechanisms: one that allows people to mitigate the episodic strengthening of repeated thoughts, and another that helps avoid looping in a repetitive thought. Self-reported efficiency in controlling unwanted thoughts correlated with both reactive thought control, and the episodic strengthening of repeated thoughts. Second, I will present data from a large study investigating formal thought disorder as a dimensional construct. In this case, the SMP can be used to show that atypical associations in individuals high on this dimension, are the result of shallow associative maps, rather than impaired control processes or effortful creativity. Finally, I will discuss potential extensions of this model to study other sequential thought and memory processes.


Isaac Fradkin
Marie Skłodowska-Curie Post-Doctoral Fellow,
Max Plank UCL Centre for Computational Psychiatry and Again research


Isaac Fradkin – Modeling the organization and control of thought in psychopathology