The beatable mind: cognitive effects of music tempo on human decision-making, a model-based approach

AGUSTÍN PÉREZ SANTANGELO; JOAQUIN NAVAJAS; MARIANO SIGMAN; MARÍA JULIANA LEONE

Conferencia; 52nd Society for Mathematical Psychology; 2019

Decision-making is essential to life. A fundamental challenge in its study is to integrate the influence of contextual factors (i.e. stimuli unrelated to the decision alternatives) into a theoretical framework of decision to better understand how we make decisions in real life. Within these factors, background music - and its speed (tempo) - is particularly relevant given the close relation it keeps with our daily experiences. Previous studies have informally shown that music affects behavior, but results are contradictory and -thus- a detailed explanation of the effects remains elusive. Critically, a systematic and quantitative approach to study the effects of music on decision-making is still lacking and the mechanistic underpinnings remain unknown.Here, we hypothesized music would either shift decision policy to be less conservative (yielding faster and less accurate decisions) through musical mode-dependent mood modulations, or distort subjective internal timing (and thus decision timing) by compressing subjective time (speeding up decisions without compromising accuracy) through music speed (tempo)-dependent arousal modulations. To test these hypotheses, we collected ~100k choice-data from standard experimental paradigms spanning perceptual, lexical, memory-inference-based and value-based decisions. We analyzed mean decision response times and accuracy and fit them jointly to a well-established cognitive model of decision-making -Drift Diffusion model (DDM) - to dissect effects at the decision-process level. We found that music ?regardless of its tempo- generally induced subjects to decide faster while, at the same time, lowering their accuracy. These joint effects were mainly attributable to a shift in the decision policy, namely, a reduction in the amount of accumulated evidence required to decide, i.e. the evidence boundary in DDM. However, these effects were task-dependent with faster decision times in four of the five tasks, and lower accuracy in three of those four tasks. Our results argue against the timing hypothesis, and support the policy-shift hypothesis for the observed decision speed and accuracy effects.Our findings extend to our daily experiences: a common, ubiquitous and seemingly innocuous stimulus shapes our decisions, making us faster at the expense of our accuracy.