A Problem for Minimal Theory of Mind
If the slides are not working, or you prefer them full screen, please try this link.
Notes
Background
Mindreading is the process of identifying a mental state as a mental state that some particular individual, another or yourself, has. To say someone has a theory of mind is another way of saying that she is capable of mindreading.
Butterfill & Apperly (2013) constructed a minimal theory of mind. This theory describes a model of minds and actions which, if it were implemented, would enable you to track others’ false beliefs. At least within limits.
This minimal model has a signature limit: it does not enable you to track false beliefs which essentially involve a mistake about numerical identity. Such as Lois Lane’s false belief that Superman and Clark Kent are distinct people (Jerry & Joe, 1939).
Signature limits generate predictions. Automatic belief-tracking in adults, and belief-tracking in infants, are both subject to signature limits associated with minimal theory of mind.[1]
The Kovács Effect
Kovács, Téglás, & Endress (2010) established that another’s irrelevant belief can influence how quickly you can detect the presence of an object. Despite some initial doubts (Phillips et al., 2015), this finding has been widely replicated by several labs (including Wel, Sebanz, & Knoblich, 2014; Edwards & Low, 2017; El Kaddouri, Bardi, De Bremaeker, Brass, & Wiersema, 2020).
Question
Why do others’ false beliefs ever have an effect on your own actions?
Motor Mindreading Conjecture
Predictions of the Conjecture
-
In motor mindreading only, goal-tracking will manifest sensitivity to agents’ beliefs.
-
In motor mindreading only, physically constraining protagonists or participants will impair belief tracking.
This talk concerns the second prediction only.
Findings So Far
Low, Edwards, & Butterfill (2020) support the prediction: physically constraining a protagonist did impair belief tracking.
Six (2022, p. Experiment 2) did not support the prediction: physically constraining participants did not impair their belief tracking.
And the results from a study in preparation that builds on Zani, Butterfill, & Low (2020)’s balance paradigm found only suggestive evidence for the prediction.
Glossary
Tracking an attribute or thing is contrasted with computing it. Unlike tracking, computing typically requires that the attribute be represented.
References
Endnotes
In favour: Wang, Hadi, & Low (2015); Low & Watts (2013); Low, Drummond, Walmsley, & Wang (2014); Mozuraitis, Chambers, & Daneman (2015); Edwards & Low (2017); Fizke, Butterfill, Loo, Reindl, & Rakoczy (2017); Oktay-Gür, Schulz, & Rakoczy (2018); Edwards & Low (2017); Edwards & Low (2019). Against: Kulke, von Duhn, Schneider, & Rakoczy (2018) argue that although the paradigm from Low & Watts (2013) replicates, attempts to modify it to avoid confounding factors do not produce comparable results. See also Scott, Richman, & Baillargeon (2015); Carruthers (2015); Carruthers (2015a); Kampis & Kovács (2022). ↩︎