Biased by our imaginings

Abstract

I propose a new model of implicit bias, according to which implicit biases are constituted by unconscious imaginings. I argue that my model accommodates characteristic features of implicit bias, does not face the problems of the doxastic model, and is uniquely placed to accommodate the structural heterogeneity in the category of implicit bias. Finally I turn to how my view relates to holding people accountable for their biases and what we know about intervention strategies.

Beliefs and biases

“…If new evidence won’t cause you to change your behavior in response to an apparent stimulus, then your reaction is due to alief rather than belief” ( 2008b , p. 566). Others make similar arguments, e.g., Grace Helton (forthcoming), Alex Madva ( 2016 ), Neil Levy ( 2014 ), and Ema Sullivan-Bissett ( 2019 ). These arguments conclude that implicit biases are unlike beliefs, which are sensitive to relevant changes in evidence.…”

“…One may wonder: Do I really believe that men are more intelligent than women? Many say no (Gendler, 2008a , 2008b ; Levy, 2014 ; Machery, 2016 ; Sullivan‐Bissett, 2019 ). The arguments vary, but we can identify two major reasons why critics claim that implicit biases are not beliefs.…”

“…Indeed, if implicit biases are completely unresponsive to evidence one might question whether they even have a mind-to-world direction of fit (3). Philosophers who have argued that implicit biases are not beliefs defend different sorts of positive accounts of implicit bias: aliefs (Gendler, 2008a , b ), patchy endorsements (Levy, 2014 ), character traits (Machery, 2016 ), unconscious imaginings (Sullivan‐Bissett, 2019 ), etc.…”

The Rational Dynamics of Implicit Thought

“…Philosophers of mind have developed a litany of proposals for making sense of the implicit a itudes. Inter alia, it has been proposed that implicit a itudes are constituted by associations between concepts [Holroyd 2016;Byrd 2019], aliefs [Gendler 2008;Madva 2016], beliefs [Mandelbaum 2016;Egan 2011;Frankish 2016], affective tension clusters [Brownstein 2018], character traits [Machery 2016], the outputs of three separate evaluation systems [Huebner 2016], reasoning structures without corresponding states [Johnson 2020], imaginings [Welpinghus 2020;Sullivan-Bisse 2019], mental imagery [Nanay forthcoming], and sui generis cognitive states like patchy endorsements [Levy 2015] and in-between beliefs [Schwi gebel 2010]. Understanding the character and function of implicit a itudes is important, both for the philosophical project of accounting for the states that furnish the mind, and for normative projects that a empt to evaluate agents and their actions, where the influence of implicit a itudes may loom large.…”

Can imagination be unconscious?

“…Thanks to an anonymous referee for pushing me on this point.27 Perhaps there might be other relevant contexts as well-for example, were unconscious imagination to exist, it might be helpful in explaining implicit bias (seeSullivan-Bissett, 2019) and low-level mental simulation (seeSpaulding 2016). Though I don't have the space here to address these claims in any detail, I'll just briefly note that there are many other widely accepted explanations of implicit bias on offer, and the existence of low-level mental simulation is highly controversial.28 Earlier versions of this paper were presented at workshops at the University of Geneva (2018) and the University of Toronto (2019) as well as at a colloquium at Ruhr University Bochum (2019).…”