Skip to main content

Gustav Markkula's online resources

An accumulator model of detection of visually looming collision threats

Link: https://github.com/gmarkkula/LoomingDetectionStudy

MATLAB implementations of the various accumulator (and threshold) models of visual looming detection described in (Markkula et al., 2021). The code can be found in the analysis folder in the Github repository, which also includes code for non-model analyses of the dataset used in the paper (as described here). The model implementations can be found in the SimulateOneTrial_AccumulatorModel.m and SimulateOneTrial_ThresholdModel.m functions, and an example of use of these functions can be found in do_19_model_MLEModelSimulationsForFigures.m which makes use of the SimulateDataSetFromModel.m function, which in turn calls the SimulateOneTrial_*.m functions.

Markkula, G., Uludağ, Z., Wilkie, R. M., & Billington, J. (2021). Accumulation of continuously time-varying sensory evidence constrains neural and behavioral responses in human collision threat detection. PLoS Computational Biology, 17(7), e1009096. https://doi.org/10.1371/journal.pcbi.1009096