Brain regions as difference-makers

Abstract

Contrastive neuroimaging is often taken to provide evidence about the localization of cognitive functions. After canvassing some problems with this approach, I offer an alternative: neuroimaging gives evidence about regions of the brain that bear difference-making relationships to psychological processes of interest. I distinguish between the specificity and what I call the systematicity of a difference-making relationship, and I show how at least some neuroimaging experiments can give evidence for systematic difference-making.

Keywords:

• Causation
• difference-making
• fMRI
• neuroimaging
• neuroscience

Acknowledgements

Thanks to Rosa Cao, Max Coltheart, Peter Clutton, Adrian Currie, Paul Griffiths, Annelli Janssen, David Kaplan, Brendan Ritchie, Kim Sterelny, Karola Stolz, Dan Weiskopf, and Jim Woodward for helpful discussions, and to audiences in Taipei, Canberra, Sydney, and Chicago for helpful feedback on earlier drafts.

Notes

1. I take as representative peaks from Ishai and colleagues (1999) and Clarke and Tyler (2014). Studies found using Neurosynth, http://neurosynth.org, described in (Yarkoni, Poldrack, Nichols, Van Essen, & Wager, 2011).

2. There is, properly speaking, no variation between tasks on such a picture, simply variation within tasks. In that regard, the present picture resembles Sternberg’s (2011) account. However, my account is more permissive than Sternberg’s. Sternberg’s model requires only qualitative changes in performance of the same task, rather than across different tasks (2011, p. 191). The determinable–determinate structure of tasks is (in many cases) hierarchical, and so a region could make a difference between task determinates which are themselves determinables of a higher level determinate. To take a fictional example, if recognizing objects and recognizing letters are ways of recognizing things, then we can meaningfully ask which brain regions make that difference. Conversely, when tasks are not plausibly related in that hierarchical way, contrasts between them don’t make much sense in the first place: you could try to contrast (say) dancing and reading, but it’s not clear why that would be a good idea.

3. Though, contra Bennett and Hacker (2003), I do not think that localizing personal-level processes is a conceptual confusion. The cruder phrenologists might have been correct, and personal capacities might have been localizable as well. We just have very strong empirical evidence that personal-level processes do not localize, and that localization itself will be of “elementary operations” (Petersen & Fiez, 1993).

4. For actual difference-making and its important role in scientific discovery, see Waters, 2007. Note, however, that Waters’ focus on actual difference-making makes the resulting picture sensitive to the actual probability distributions of different causes (Griffiths et al., 2015), a feature that I intend to avoid.

5. See DiCarlo and Cox (2007, p. 334) for the general picture. Note that this is an idealization; non-intersecting tangledness is only possible if each point in V1 space always corresponds to a unique object category in the division scheme, which is unlikely given the role of time-dependent effects on categorization.