Experiment example (PsychoPy): Masked Priming English
Authors: Crystal Jemy, Roberto Petrosino
Description
A typical trial of a masked priming lexical decision experiment consists of the following:
a series of ###### (mask): 500 ms
a prime stimulus (in our case, a word): variable duration (in the original script, 33 ms). Crystal’s project: 50 ms. Roberto’s project: 33 ms.
a TARGET stimulus (in our case, an all-caps word): 2500 ms (or until participant’s response). The participant is asked to decide whether the TARGET is a word or not as fast as possible by pressing two buttons; in the original script, a keyboard was used, so the j and f keys were used for word and non-word decision, respectively.
The loop of #####-prime-TARGET triplets will be fed from a csv file containing all pairs to be presented in a single recording. Typically, an experiment will have at least two different stimulus lists (but there may be more), and each participant is assigned one at the beginning of the recording. This is to guarantee proper counter-balancing across participants.
Trigger setup
We plan to use a subset of the 8 digital trigger lines (BNC / parallel port channels) set to individual single-line pulses. All trigger pulses will be sent exclusively upon TARGET word onset and synchronized with the photodiode signal to ensure sub-millisecond precision.
Project 1: Crystal’s Study ~ Latin Square design with 5 experimental conditions.
Trigger Lines (4 lines): Each condition triggers a dedicated single line pulse upon TARGET onset:
Line 1: Identity
Line 2: Morphological
Line 3: Orthographic
Line 4: Unrelated
Lne 5: Non-word
Project 2: Roberto’s Study ~ 2 × 6 design crossing prime-target relatedness (Related vs. Unrelated) with condition type (Identity, Morphological, Opaque, Orthographic, Semantic, Non-word).
Trigger Lines (7 lines total):
Line 1 (Relatedness Flag): Pulsed only for Related trial pairs.
Lines 2–7 (Condition Flags): Dedicated single-line pulse for each target condition:
Line 2: Identity
Line 3: Morphological
Line 4: Opaque
Line 5: Orthographic
Line 6: Semantic
Line 7: Non-word
Trial-level metadata (e.g., keypresses, RTs, item IDs, stimulus characteristics) will be recorded in the presentation log files and merged post-hoc during MEG preprocessing using trial indices.
Eyetracking
The same events (prime, TARGET, and response) will also need to be set up with the eyetracker.
Additional technical issues (in order of importance)
Response Hardware Selection. The lab provides both multi-button and two-button response boxes. We need to determine which model is optimal for our experimental setups.
Experimental Breaks. Participants will be given periodic rest breaks during testing, with the total number of breaks determined by the specific trial count of each experiment.
Data Collection Interface. While a full graphical interface (GUI) makes running scripts fast and user-friendly, it carries potential trade-offs in stability and bug risks. A strong alternative is a lightweight startup shell (or prompt dialog) that allows the experimenter to quickly (i) enter the participant ID and (ii) select the stimulus list to present.
Automated Script Generator (GUI/Tool). Longer-term, it would be helpful to develop a standalone utility script that allows students to input custom experimental parameters (e.g., prime duration, break frequency, stimulus file naming) and automatically generate a pre-configured masked priming script.