Gradient boosted decision trees reveal nuances of auditory discrimination behaviour

• Published in: bioRxiv
• Main Authors: Griffiths, Carla Seoyun, Lebert, Jules M, Sollini, Joseph, Bizley, Jennifer K
• Format: Paper
• Language: English
• Published: Cold Spring Harbor Cold Spring Harbor Laboratory Press 28.02.2024; Cold Spring Harbor Laboratory
• Edition: 1.3
• Subjects: Auditory discrimination; Decision making; Decision trees; False alarms; Frequency; Hemispheric laterality; Mustela; Neuroscience; Psychophysics; Reaction time task; Streaming; pitch; behavioral data analysis in neuroscience; Shapley Additive Explanations; ferret; auditory scene analysis
• ISSN: 2692-8205; 2692-8205
• DOI: 10.1101/2023.06.16.545302

Animal psychophysics can generate rich behavioral datasets, often comprised of many 1000s of trials for an individual subject. Gradient-boosted models are a promising machine learning approach for analyzing such data, partly due to the tools that allow users to gain insight into how the model makes predictions. We trained ferrets to report a target word's presence, timing, and lateralization within a stream of consecutively presented non-target words. To assess the animals' ability to generalize across pitch, we manipulated the fundamental frequency (F0) of the speech stimuli across trials, and to assess the contribution of pitch to streaming, we roved the F0 from word token-to-token. We then implemented gradient-boosted regression and decision trees on the trial outcome and reaction time data to understand the behavioral factors behind the ferrets' decision-making. We visualized model contributions by implementing SHAPs feature importance and partial dependency plots. While ferrets could accurately perform the task across all pitch-shifted conditions, our models reveal subtle effects of shifting F0 on performance, with within-trial pitch shifting elevating false alarms and extending reaction times. Our models identified a subset of non-target words that animals commonly false alarmed to. Follow-up analysis demonstrated that the spectrotemporal similarity of target and non-target words rather than similarity in duration or amplitude waveform was the strongest predictor of the likelihood of false alarming. Finally, we compared the results with those obtained with traditional mixed effects models, revealing equivalent or better performance for the gradient-boosted models over these approaches.Competing Interest StatementThe authors have declared no competing interest.Footnotes* This version of the manuscript has been revised to further link false alarms to acoustic features of non-target sounds, and to provide a more reader-friendly explanation of the methods used.