Streamlining experiment design in cognitive hearing science using OpenSesame

Abstract

Auditory science increasingly builds on concepts and testing paradigms originated in behavioral psychology and cognitive neuroscience - an evolution of which the resulting discipline is now known as cognitive hearing science. Experimental cognitive hearing science paradigms call for hybrid cognitive and psychobehavioral tests such as those relating the attentional system, working memory, and executive functioning to low-level auditory acuity or speech intelligibility. Building complex multi-stimuli experiments can rapidly become time-consuming and error-prone. Platform-based experiment design can help streamline the implementation of cognitive hearing science experimental paradigms, promote the standardization of experiment design practices, and ensure reliability and control. Here, we introduce a set of features for the open-source python-based OpenSesame platform that allows the rapid implementation of custom behavioral and cognitive hearing science tests, including complex multichannel audio stimuli while interfacing with various synchronous inputs/outputs. Our integration includes advanced audio playback capabilities with multiple loudspeakers, an adaptive procedure, compatibility with standard I/Os and their synchronization through implementation of the Lab Streaming Layer protocol. We exemplify the capabilities of this extended OpenSesame platform with an implementation of the three-alternative forced choice amplitude modulation detection test and discuss reliability and performance. The new features are available free of charge from GitHub: https://github.com/elus-om/BRM_OMEXP .

Keywords: Cognitive hearing science; Experiment building; Experiment design; OpenSesame.

Conflict of interest statement

The authors declare that they have no conflicts of interest.

© 2022. The Author(s).

Figures

Fig. 1

Visual cues used for the 3-AFC-AMDT. Three sound stimuli are sequentially played and each stimulus is associated to one of the red rectangles displayed on a screen in front of the participant

Fig. 2

3AFC AMDT logic translated to the experiment structure

Fig. 3

3AFC AMDT experiment sequence: it consists of a collection of actions (plugins) which will be run in a sequential order. Calibration whitenoise, new lsl start 1, new adaptive init, stimulus and change represent the new plugins introduced in this manuscript

Fig. 4

Inline script to generate the sound stimuli

Fig. 5

Mixer plugin

Fig. 6

Adaptive plugin

Fig. 7

Automatic answer generated using an inline script plugin

Fig. 8

The psychometric function that was used to generate the synthetic response: the orange circle represents the midpoint of the curve and the blue dashed lines represent the expected convergence [depth] value our 3-down 1-up adaptive procedure

Fig. 9

LSL start plugin

Fig. 10

Visual representation of the sound stimuli and the specific moments at which the LSL messages are recorded. The Audio Mixer plugin plays three different audio stimuli sequentially, depicted as striped boxes. The orange rectangles represent the silent breaks between the three sound stimuli. For each audio stimulus, the lsl message plugin saves an LSL marker for the time in which the audio files starts and stops to be played

Fig. 11

3-down-1-up procedure results for the first simulated test participant and for the first of the four conditions (frequency = 400 Hz and modulation frequency = 5 Hz)

Fig. 12

Platform behavior validation of the adaptive plugins. The x-axis represents the 40 tested conditions, four combinations of frequencies, and modulation frequencies for the ten synthetic subjects. The y-axis shows the depth value range. The black line shows the depthmean(j), where j goes from 1st to the 40th tested condition, and the error bars represents the depthstd(j). The dashed green line gives the expected convergence value (=-7.3622 dB)

Fig. 13

The acquired LSL sound stream is represented: on the top, the whole sound stream consisting of 144 AM tones is shown; on the bottom, the sounds related to one trial (3x3 audio files) are depicted

Fig. 14

Random error distribution for the ten simulated subjects responses