Matthew Botvinick - 54th Bartlett Prize Lecture (2026)

Matthew Botvinick - 54th Bartlett Prize Lecture (2026)

🎙 Matthew Botvinick 👥 391 📅 March 3, 2026 ⏱ 56 min 👁 318 📄 expert opinion 🧭 2026-08-16
Available in: English (current) Français

Keywords

cognitive controlhabit systemexecutive controlcognitive effortlabor supply theory

Summary

In this Bartlett Prize Lecture, Matthew Botvinick reflects on his career and presents a synthesis of his work on cognitive control. He begins by describing two seemingly incompatible strands of his PhD research: one on conflict monitoring and the executive control system, and another on modeling sequential action with unstructured recurrent neural networks. He highlights the tension between the classical view of distinct habit and control systems and the eliminativist connectionist approach that shows such distinctions may not be computationally necessary. He then introduces a second riddle: why cognitive effort is aversive, presenting empirical evidence from his lab’s demand selection task and labor supply theory experiments. Finally, he proposes that the cost of cognitive control and the existence of separate systems may be explained by the need for generalization and transfer, suggesting that specialized control systems evolved to enable flexible behavior across diverse contexts. The lecture integrates personal narrative with scientific content, offering a unique perspective on the field.

159 words

Critical Evaluation

Value of the Information & Strength of the Argument

The lecture provides valuable insights into the theoretical foundations of cognitive control, bridging classical cognitive psychology and modern neural network modeling. Botvinick’s argumentation is coherent and well-supported by empirical evidence, including his own studies and those of others. He effectively raises important questions about the necessity of architectural distinctions and the nature of cognitive effort, offering a compelling computational perspective. The use of labor supply theory to explain mental effort is particularly innovative and adds depth to the discussion.

Scientific Rigor, Source Quality, Title Accuracy

Botvinick demonstrates scientific rigor by grounding his talk in established research, referencing key studies and models (e.g., Cohen’s Stroop model, Cooper and Shallice’s hierarchical models, Harlow’s learning-to-learn task). He cites his own published work and that of collaborators, providing a credible basis for his claims. The title accurately reflects the content, as it is a prize lecture summarizing his research contributions. The talk is well-structured and maintains a high level of scientific integrity.

167 words

Title / Content Match

The title accurately reflects the content: a prize lecture by Matthew Botvinick, covering his research on cognitive control and dual-process cognition.

Quality & Reliability

8/10

The lecture is delivered by a leading expert in cognitive neuroscience and AI, with references to established research and his own published work. The content is well-structured and grounded in empirical evidence, though it is a personal synthesis rather than a systematic review.

Key Moments

Cited Sources

  • Cohen, J.D., et al. (1990). On the control of automatic processes: A parallel distributed processing account of the Stroop effect. — Mentioned as the influential model of the Stroop task.
  • Norman, D.A., & Shallice, T. (1986). Attention to action: Willed and automatic control of behavior. — Referenced as a classical model distinguishing automatic and control systems.
  • Cooper, R., & Shallice, T. (2000). Contention scheduling and the control of routine activities. — Mentioned in the context of hierarchical models of action.
  • Harlow, H.F. (1949). The formation of learning sets. — Introduced the learning-to-learn task.
  • Collins, A.G.E., & Frank, M.J. (2013). Cognitive control: A role for the prefrontal cortex in learning and decision-making. — Referenced as addressing abstract task rules and cognitive control.
  • Westbrook, A., & Braver, T.S. (2015). Cognitive effort: A neuroeconomic approach. — Mentioned for a study putting a dollar value on cognitive effort.

Concurring Sources

  • Botvinick, M.M., et al. (2001). Conflict monitoring and cognitive control. — Botvinick's own work on conflict monitoring, consistent with the lecture's themes.
  • Botvinick, M.M., & Plaut, D.C. (2004). Doing without schema hierarchies: A recurrent connectionist approach to normal and impaired routine sequential action. — Directly supports the eliminativist approach discussed in the lecture.

Dissenting Sources

  • Miller, E.K., & Cohen, J.D. (2001). An integrative theory of prefrontal cortex function. — This influential theory emphasizes the necessity of a distinct prefrontal control system, which contrasts with the eliminativist view presented.

Contribution & Novelties

The lecture offers a novel synthesis of two seemingly contradictory research traditions in cognitive control: the classical view of distinct habit and control systems and the eliminativist connectionist approach. Botvinick proposes that the cost of cognitive effort and the existence of separate systems may be explained by the need for generalization and transfer, providing a fresh computational perspective. This integrative framework could inspire new research directions.

Pour aller plus loin :

120 words

Radar Profile

The radar profile shows high scores in quality of information and global reliability, reflecting the expert's credibility and well-supported arguments. The lower score in technical level suggests the content is accessible to a broad audience, while still providing depth for specialists.

Reliability 8/10

💬 No comments were provided for analysis.