Valence is among the principle constructs used to explain animal behavior; we tend towards things we like, away from what we don’t. Remarkably, not much is known about how valence effects processes upstream of decision-making–e.g. memory and perception. In the hopes of characterizing the relationship between valence and perceptual processing, Schechtman et al. 2010 showed differential patterns of generalization in a perceptual learning paradigm, as a function of valence. After a learning protocol where an auditory cue was positively reinforced, subjects were better at discriminating between that from similar cues; subjects performed worse at this task when the tone was negatively reinforced.
My hope is to build upon this core finding–differential generalization as a function of valence–in order to study how these perceptual processes interact with memory. By replicating this study in an online sample, I would be well positioned to iterate through paradigms, eventually converging on a paradigm that will enable me to explore the relationship between valence and memory.
This experiment is well suited for adaptation in an online setting: The stimuli are auditory tones (e.g. around 300, 500, 700, Hz tones presented for 200ms); responses are key presses (3 keys); the protocol ensures that subjects who attend to each trial are able to minimize losses and earn rewards on top of their base pay–that is, the experimental design incentives subjects to remain engaged, and the pattern of behavioral evidence will allow us to identify subjects who were not engaged.
The experiment consists of two parts, an acquisition stage and a generalization stage. In the acquisition stage, subjects are told that in each trail they will hear one of three tones, and that each tone has a corresponding key, which must be pressed with 2.5 seconds after tone presentation. One tone-key pairing will gain them money (the “positive tone”); one tone-key pair will keep them from loosing money (the “negative tone”); the last tone results in neither loss nor reward (neutral valence), regardless of what key is pressed. Subjects must learn via trial and error which tone-key pairings lead to the desired outcomes; gaining rewards or minimizing losses.
In the generalization stage, subjects are presented with the original three tones, as well as range of similar (e.g. original tones +- 5, 10, … Hz) and dissimilar tones (e.g. 50 Hz). If the tone is either the positive or negative tone, subjects are instructed to press the key that corresponded to that tone in the acquisition stage. Otherwise, they are instructed to press a third key. Subjects are rewarded for the correct response (positive key, negative key, or novel key, accordingly) and penalized for non-responses.
While the stimuli, responses, and incentives of this task are well suited for online experiments, one of the central challenges in this replication will be calibrating the monitary incentives and disencentives.