Data and code from: Beyond masking: Cognitive limits on mate choice
Description
Mate choice poses a demanding and high-stakes information-processing challenge: receivers must detect, remember, and compare mating signals across multiple potential mates. Strong preferences in simple contexts often weaken when receivers encounter numerous signalers, reshaping sexual selection. A potential mechanism driving this is energetic masking, when overlapping signals saturate sensory channels, hindering receivers’ ability to detect and discriminate among mates. Alternatively, choice overload, when the cognitive demands of evaluating too many options exceed an individual’s processing capacity, could reduce decision performance. Although energetic masking is well documented, we know little about choice overload in nonhumans. Here, we investigated both mechanisms in Cope’s gray treefrogs (Hyla chrysoscelis) using playback experiments that independently manipulated the number of calling males and the presence of background chorus noise. Females chose between a longer “target” call and one, three, or seven shorter “distractor” calls in the presence and absence of background noise. As the number of non-overlapping calls increased, females were substantially less accurate in selecting the target, took longer to decide, and more frequently deferred choice altogether. Noise only modestly affected discrimination, although in the opposite direction than expected. These results demonstrate that cognitive limits on information processing can subvert mate choice, even in the absence of sensory interference. Our findings highlight choice overload as a novel and underappreciated constraint on sexual selection. Accounting for cognitive limitations may help explain persistent variation in sexual traits despite apparently strong directional selection (“paradox of the lek”) and help refine models of selection dynamics in complex social and acoustic environments.
Files
Steps to reproduce
We presented receivers with a choice between one longer-duration ‘target’ call and one, three, or seven shorter-duration ‘distractor’ calls, and we examined how the number of call options and background noise influenced female gray treefrog phonotaxis. Calls were presented sequentially, such that only one call was broadcast at any given moment, eliminating temporal overlap (Figure 2A,B). We measured three behavioral responses: 1) choice accuracy, the probability of choosing the target call; 2) latency to choose, the time elapsed before making a choice; and 3) choice deferral, the probability of not responding during the allotted time. Each female was tested across all combinations of option number (two, four, or eight) and background noise condition (quiet or noise), yielding six unique treatment combinations. Between subjects, we randomized the location of the target call speaker. In the eight-option tests, all remaining speakers broadcast distractor calls. In the two- and four-stimulus tests, we randomized which speakers were used and which remained silent in order to control for potential effects of spatial relationships between target and distractor locations. Additionally, we randomized the temporal order of call onset so that the target call could occur in any order within the presentation sequence, from first to last. Because we wished to control the time elapsed between any two subsequent call presentations to account for any possibility of forward masking 58, we did not allow empty “timeslots”; instead, in the 2- and 4- choice tests, only the first two and four timeslots were used, respectively. This means that, in eight-choice tests, calls were presented sequentially and repeated in order with equal-length pauses between consecutive calls (e.g., D1 D2 D3 T4 D5 D6 D7 D8), whereas in two- and four-choice tests only the first two or four temporal positions were filled, and the remaining positions were left silent before repeating the sequence (Fig. 2A and S1). Each female was tested with two randomized combinations of spatial and temporal relationships between the target and distractor(s), resulting in 12 trials. Fifty-two females completed all 12 trials while eight females completed a subset (ntrials= 675; nfemales= 60; Table S5).
Institutions
- University of Tennessee at KnoxvilleTennessee, Knoxville
- Smithsonian Tropical Research InstitutePanamá, Panama City