Dr Adam Linson (University of Oxford & University of Edinburgh) on "Shifting perspectives: From veridical to reliable perception across physical and aesthetic contexts".
Abstract: In this paper, I introduce a general criterion of the “reliability” of perception (relative to some aim) that generalizes across time scales, which I suggest is an effective replacement for the traditional sense of veridical perception. My argument consists of three steps: (1) I argue that, in a wide number of cases, veridical and reliable perception are co-extensive with each other, which I claim has to do with scientific systems and technologies of measurement (that ground veridicality) and the physical level of description (that is the target of scientific measurement, and also relates to reliability in physically-dependent cases, e.g., crossing the road safely); (2) I introduce Pylyshyn’s (1999) account of early perception (defined only for vision), which suggests that most early vision -- due to the co-evolution of organism and niche -- approximates veridical perception (without guaranteeing it). If he is right, we might expect an analogous sense of “early audition”, but I will show that this concept poses unresolvable difficulties for the notion of veridical perception. I also compare his position to that of Lupyan (2015), to raise further issues with veridical perception using the framework of predictive processing (PP). I then argue that (3) when considering cases of cultural aesthetic experience, there is a useful way of conceptualising aesthetic production (consistent with PP) such that (a) the notion of veridical perception becomes untenable and (b) the notion of reliable perception as I define it functions consistently across the cases I present in (1)-(3). My central example in (3) is musical improvisation, which brings into especially sharp focus the issues of veridicality and reliability across physical and aesthetic contexts, due to its cognitive overlap with ordinary, unplanned environmental interaction.