🧠 Serial Dependence in Time Perception under Uncertainty

This repository contains the full implementation of a behavioral experiment exploring serial dependence in time perception under varying levels of sensory uncertainty. The experiment is built with PsychoJS and ready for deployment via Pavlovia.

---

📋 Overview

This experiment investigates how uncertainty modulates serial dependence in a time reproduction task using Random Dot Kinematograms (RDKs). Participants observe a motion stimulus and reproduce its duration. Coherence is varied to introduce high vs. low uncertainty conditions.

---

🧪 Experimental Procedure

🔄 Block Design

• The task includes multiple blocks, each with 54 trials.
• The first block is a practice block.
• Before each block, a prompt appears: ➤ "Block (X / N) will start. Please press 'S' to continue."

🔁 Trial Flow

Each trial consists of the following stages:

1. Masking Phase (Mask_2)
   A low-coherence RDK is presented to reduce aftereffects.

2. Encoding Phase (M2)
   A directional RDK is shown for a randomized duration (e.g., 0.8–1.6 s). Direction is randomized per trial.

3. Cue Phase (Cue)
   A cue ("T") appears, signaling the participant to start reproduction.

4. Reproduction Phase (Response)
   Participants hold the ↓ key to reproduce the observed duration. Duration of the key press is recorded.

5. Feedback Phase (Feedback_Time)
   Five colored circles give visual feedback based on the relative error between target and reproduced duration.

---

🔧 Technical Highlights

This experiment includes several key features to ensure precision and device consistency:

• ✅ Screen Calibration
  Participants resize a reference image (e.g., credit card) to match real-world size. This enables accurate pixel-per-cm scaling.

• ✅ Frame Rate Compensation
  The frame rate of the participant's display is measured, and dot movement speed is automatically adjusted to maintain consistent motion in cm/sec across devices.

• ✅ RDK
  A custom JavaScript-based RDK engine supports:

• Coherence control

• Noise modes: "mask", "inertial", "replace", "walk"

• Dot scaling, speed control, and dynamic pixel mapping

---

🗃 Data Logging

Each trial logs the following information (to .csv):

Column	Description
Block_id	Block index
trial_in_block	Trial number within the block
coherence	Motion coherence level (uncertainty)
direction	RDK motion direction (degrees)
TimeDur	True stimulus duration (in seconds)
keyDuration_global	Participant’s reproduced duration
feedback_ID	Feedback index (1–5)
relative_error	Signed error between true & reproduced
x_scale, y_scale	Screen scaling calibration values

---

🌐 Deployment Guide

To run the experiment online using Pavlovia:

1. Upload the following files to your Pavlovia Git project:

• Serial_Dependence_U.psyexp
• Serial_Dependence_U.js
• `resources (containing images, CSVs, and the custom RDK script)

2. Sync the project using PsychoPy or Git.

3. Share the experiment link with participants.