Comparing visual and auditory cued eyeblink conditioning in humans
Description
This dataset contains anonymized single-trial data from two human eyeblink conditioning experiments. The study examined whether conditioned responses can be detected and timed using pupillometry in addition to the standard orbicularis oculi EMG measure, and whether response incidence and onset differ across stimulus modalities, trial types, and learning phases. The data include habituation, acquisition, and extinction trials, with auditory and visual conditioned stimuli in Experiment 1 and two visual stimulus conditions in Experiment 2. For each trial, conditioned-response incidence is coded as 0 or 1, and onset latency is reported in milliseconds relative to conditioned-stimulus onset. Empty onset cells indicate that no conditioned response was detected. The dataset can be used to compare pupil- and EMG-based responses across participants, sessions, phases, and paired versus unpaired trials.
Files
Steps to reproduce
Data were collected using a classical delay eyeblink conditioning protocol administered with Presentation software. Conditioned stimuli lasted 550 ms and were either auditory tones or colored visual stimuli. A 100 ms air puff directed to the right eye served as the unconditioned stimulus, beginning 450 ms after conditioned-stimulus onset and co-terminating with it. Each session included habituation, acquisition with 70% reinforcement, and extinction. Experiment 1 included a screening session followed by auditory and visual task sessions, whereas Experiment 2 included two visual task sessions. Orbicularis oculi EMG was recorded at 2 kHz using a BIOPAC MP160 system, and pupil and blink data were recorded at 500 Hz using an EyeLink 1000 eye tracker. EMG responses were identified semi-automatically and visually verified, while pupillometry-based responses were manually evaluated in MATLAB using the eye tracker’s blink and pupil-size signals. Responses between 150 ms after conditioned-stimulus onset and unconditioned-stimulus onset in paired trials, or conditioned-stimulus termination in unpaired trials, were classified as conditioned responses. The shared dataset contains the resulting anonymized trial-level response classifications and onset latencies.