Abstract

Abstract:

One of the tasks of philosophy is to offer methodological critiques of other disciplines. In this journal club, we will discuss two papers from within neuroscience that dissect our tools and techniques. By examining the philosophical assumptions that underlie these methods, the papers aim to shed a light on what they see as prominent problems within modern neuroscience. In particular, both papers are concerned with the issue of what is means to understand a cognitive system, and whether we have the capacity to generate that understanding.

The first paper, “Neuroscience Needs Behaviour: Correcting a Reductionist Bias” (Krakauer et al. 2017) examines the role of behavioural data in neuroscientific analysis. The authors advance a thesis that neuroscientific analysis focusing on constituent elements of the nervous system, such as individual neurons and circuits, is insufficient to generate a rich understanding of the brain and the computations it performs. Instead, an understanding-oriented analysis much begin with the repertoire of natural behaviours the nervous system is capable of producing.

The second paper, “Could a neuroscientist understand a microprocessor?” (Jones and Kording 2017) asks whether standard neuroscientific techniques are capable of generating the sort of understanding that the discipline seeks. To test these standard techniques, the authors apply them to a computational system (a microprocessor) and ask whether they are sufficient to uncover the computations being performed. Finding them wanting, they then ask what these techniques must be supplemented with to give us a toolkit for understanding computational processes.

References:

[1] “Neuroscience Needs Behaviour: Correcting a Reductionist Bias” (Krakauer et al. 2017)

[2] “Could a neuroscientist understand a microprocessor?” (Jones and Kording 2017)

Zoom information:

https://eng-cam.zoom.us/j/84204498431?pwd=Um1oU284b1YxWThObGw4ZU9XZitWdz09 Meeting ID: 842 0449 8431 Passcode: 684140