Project 8: Response Times Experiment

Description

In this project, you will design a response time experiment, implement a Java program that will collect data, and then conduct a pilot experiment to test your experimental setup.

Response times

The measurement of response times can be used to investigate a wide range of learning, psychological, or social phenomena. In Lab 9 we used ResponseTimeExperiment.java to measure response times to addition problems between integers in the range 0...99. A program like that can be used to test response times of children learning addition, possibly as part of a game. Analysis of response times can sometimes provide support to theories of cognition or shed light on questions about associations or bias that are difficult to address in any other way -- see, for example, scientific studies in decision problems or letter recognition.

Part 1: Design an experiment and implement a java program to conduct the experiment Is there a research question involving response times that you would find interesting? The task involved can be something simple, similar to the addition problem used in the example (please do NOT use that!). It can be something even simpler, like how long it takes for a person to type 'a' or 'b' in reaction to a number that is, say, positive or negative. It can be a question related to one of your other courses. It can be anything about which you are curious, but should not be too complicated to implement in a java program. The task may involve answering a question for which there is a right/wrong answer, or it may be a judgment or decision problem for which there is no right or wrong answer, or it might even be a task where the subject must simply react when they see something - for example the letter R hiding in a random position among a bunch of B's. See some references given in the introduction for how reaction times for some very simple tasks can be used to explore important questions of relevance to human cognition or larger social issues. Each run of the program should conduct the experiment with a single subject. The program can be run repeatedly to obtain data from different subjects. Your implementation needs to be easily adaptible to conduct any number of trials for each subject, by defining a constant at the beginning of the code:

int NUM_TRIALS = 4; 

NUM_TRIALS can then be used to count trials and to compute results. The constant makes it easy to adapt the program to perform a different number of trials.

Important notes:

• Use a for-loop to run the four (or more) trials, collecting response times and any other data relevant to the task (eg, number of correct responses)
• Remember to ask the subject's name (or you can use some kind of id # if you are doing a double-blind study) and to issue clear instructions for the task
• Whatever question or task you decide to use, it is important that the individual questions be generated randomly (as, for example, in the lab 9 experiment). The program should NOT have a list of set questions that are asked every time it runs -- the questions should change with different runs.

Part 2: Conduct a pilot study and write up the results Gather some friends (at least 5) to serve as experimental subjects in order to conduct a pilot experiment. * Ask each person to run the program and record their data in a spreadsheet. * Ask each person to comment on your program and record their observations for your report. * Analyze the results of the pilot experiment and include in your project report.

Challenge for the bored This is an open-ended project, where there are many opportunities to challenge yourself. Here are some ideas:

• You can use JOptionPane for the interaction.
• The task may involve additional graphical display, for example, an image. The subject might be asked to click "yes" or "no" depending whether the text displayed or the image has certain properties (for example, is displayed on a certain color background or to the left of the caption, or depicts an animal or a person, etc). Be sure to choose a task that you can handle when it comes to implementing it in a Java program and only use images if you enjoy implementing GUIs.
• Using a question bank can be fun, but you will need to use some of the material that we have not yet covered in this course, such as arrays of Strings (we will be learning about this next week). If you have a bunch of questions stored in such an array, you can generate an index randomly to pick out questions to ask. It can be a bit tricky to avoid repeating questions though!
• You can automate the entire experiment and analysis of the results, creating an outer loop that handles the experimental subjects and generates the report (instead of manually inputting the information into a spreadsheet).