Perceptual change-of-mind decisions are sensitive to absolute evidence magnitude

An initial ‘snapshot’ of sensory information biases the likelihood and speed of subsequent changes of mind

“…For example, when continuous response measures are employed, participants may adopt different strategies because movements can be initiated before a strict decision has been finalized, and because they may be weighing up effort costs which evolve over time [ 7 , 8 , 23 ]. However, the general pattern of changes of mind we observed in this study, and in another recent study, which also employed a binary response measure [ 22 ], mirrors those observed in studies employing continuous response measures [ 3 , 5 , 7 , 8 ]. In particular, across all these studies changes of mind were more common following errors and were driven by stereotyped shifts in stimulus evidence.…”

“…This was to maximise the number of useable trials in the psychophysical reverse correlation analyses. Because the dynamics of the decision process change if the stimulus values are changed, even if just their absolute values are manipulated while their difference is held constant [e.g., 21 , 22 ], each additional stimulus condition would have effectively halved the number of useable trials. However, one concern which stems from this decision is that participants may have adopted a ‘detection’ rather than ‘discrimination’ strategy.…”

Magnitude-sensitive reaction times reveal non-linear time costs in multi-alternative decision-making

“…Rederiving optimal policies to account for geometric (Marshall, 2019) or general multiplicative (Steverson et al, 2019) costs of time qualitatively changes them in the binary decision case, introducing magnitude-sensitive reaction times (Marshall, 2019;Steverson et al, 2019). However, disentangling these from nonlinear subjective utility is challenging, and cannot be excluded as an explanation for previous results (Teodorescu et al, 2016;Pirrone et al, 2018a;Steverson et al, 2019;Zajkowski et al, 2019;Turner et al, 2019;Bhui, 2019;Smith and Krajbich, 2019;Pirrone and Gobet, 2021).…”

An initial ‘snapshot’ of sensory information biases the likelihood and speed of subsequent changes of mind

“…For example, when continuous response measures are employed, participants may adopt different strategies because movements can be initiated before a strict decision has been finalized [22]. However, the general pattern of changes of mind we observed in this study, and in another recent study, which also employed a binary response measure [21], mirrors those observed in studies employing continuous response measures [3,5,7,8]. In particular, across all these studies changes of mind were more common following errors and were driven by stereotyped shifts in stimulus evidence.…”

“…This was to maximise the number of useable trials in the psychophysical reverse correlation analyses. Because the dynamics of the decision process change if the stimulus values are changed, even if just their absolute values are manipulated while their difference is held constant [e.g., 20,21], each additional stimulus condition would have effectively halved the number of useable trials. However, one concern which stems from this decision is that participants may have adopted a 'detection' rather than 'discrimination' strategy.…”