Session 2

In this session, we’ll continue developing our understanding of cognitive experiments, focusing on the lexical decision task and exploring the use of feedback, counterbalancing, and nested lists. Additionally, we’ll introduce the first task submission involving the Flanker Task.

Developing a Lexical Decision Task

Trial Structure

Each trial in the lexical decision task involves the following sequence:

  • Fixation: Displayed for 1000 ms
  • Probe: Displayed for 2000 ms

The key goal is to understand what elements need to be fixed (e.g., fixation cross, display duration) and what should vary (e.g., the word presented, correct response).

Adding Blocks

To add more complexity, we can divide the trials into blocks, with breaks in between.

Homework

How would you add a practice block to the task?:::

Adding Feedback to Trials

  • Add a FeedbackDisplay object as the last event in your trial procedure.
  • Explore the feedback’s default options (e.g., states, duration).
  • Use the “Input Object Name” to base the feedback on a specific object (e.g., correct response).
Homework

How would you add feedback only to the practice block?

Adding All Trials

There are different ways to add trials: - If the number of trials is small, you can manually enter the conditions. - For a large number of trials, use nested lists. This involves: - Reducing the trial list to the number of conditions. - Creating a new list for each condition, which contains the trial exemplars.

Tip

Keep at least two rows per condition to avoid predictable sequences.

Counterbalancing Responses

To prevent response bias, you can counterbalance responses across participants. For example: - Participant 1: word = “k”; non-word = “l” - Participant 2: word = “l”; non-word = “k”

First Task Submission: The Flanker Task

Deadline: October 20th

In this task, you will explore the effect of congruence on response inhibition using the Flanker Task, which assesses the ability to suppress inappropriate responses.

Key Elements of the Flanker Task

  • Congruent Trials: Where flankers and the target match (e.g., HHHHH).
  • Incongruent Trials: Where the flankers and the target do not match (e.g., SSHSS).

Sequence of Events per Trial

  • Fixation (150 ms)
  • Probe (80 ms)
  • Response Window (800 ms)
  • Feedback (300 ms)

Task Design

The experiment will follow a 2 (between-participants) x 2 (within-participants) design: - Between-participants factor: Proportion of congruence (50% vs 80% incongruent trials). - Within-participants factor: Congruence (congruent vs incongruent trials).

Each participant will complete 80 trials in total, equally distributed across the conditions.

Probe List

You must use nested lists for random probe selection, ensuring that all probes are used the same number of times across trials.

Congruent Incongruent
AAAAA AABAA
BBBBB BBABB
XXXXX XXYXX
YYYYY YYXYY

Extra Points

  • Counterbalance the proportion of congruence between participants (e.g., group 1 = 50%, group 2 = 80%).
  • Run three “participants”, merge their data, and compute the mean reaction time (RT) and accuracy (ACC) for each condition.

Please upload your completed task to PRADO or send it via email. Late submissions will incur a 20% penalty.

Consider the following questionsn:

Trial Structure and Experimental Design

  • What are the advantages of using nested lists when designing experiments with a large number of trials or conditions? How does this affect the flexibility and scalability of your experimental design?
  • Why is it important to balance the number of trials across conditions, and what issues could arise if this balance is not maintained?

Feedback in Cognitive Tasks

  • How does adding feedback in a cognitive task, particularly in a practice block, help participants perform better in the actual trials? What are the potential downsides of providing feedback in all blocks?
  • Reflect on the role of feedback in learning and task performance. Should feedback always be included in experiments, or are there scenarios where it might interfere with data collection?

Counterbalancing and Bias

  • Why is counterbalancing critical in cognitive tasks, such as the lexical decision task or the Flanker task? What would be the consequences of failing to counterbalance responses across participants?
  • How does counterbalancing responses improve the internal validity of an experiment? Can counterbalancing introduce any unintended complexities in data analysis?

The Flanker Task

  • In the context of the Flanker task, how does the proportion of congruent versus incongruent trials affect response inhibition? Why might a higher proportion of incongruent trials make it easier for participants to control inappropriate responses?
  • Reflect on how varying the proportion of congruent and incongruent trials could affect the interpretation of reaction time (RT) and accuracy (ACC) results. How might these effects differ across participants with varying cognitive control abilities?

Experimental Controls and Variability

  • What are some potential challenges in designing experiments with both within-participants and between-participants factors, as seen in the Flanker task? How can you minimize participant variability while maintaining the experimental design’s integrity?
  • Why is it important to control for the sequence and timing of stimuli presentation, and how do randomization and counterbalancing help address these issues?

Below, you can find the slides from this session: