Lab 5: Testing Fitts’ Law

In your text in Chapter 6 on pages 163-168, Schmidt and Wrisberg discuss the idea of a speed-accuracy trade-off in rapid continuous movements. The idea is that if a performer moves too quickly they will sacrifice accuracy. Or said in a different way, if performers want to improve their accuracy they will need to slow down their movements. This trade-off is known either explicitly or implicitly by most skilled performers and is used as a movement strategy on a regular basis. For example, if you are filling out a document and there is a small box for your signature, people will slow down to ensure that the signature fits neatly in the box without touching the lines. People move slowly when attempting to thread a small needle. A careful driver will automatically slow down when driving through a narrow country road and speed up when the highway widens.

As it relates to Fitts’ tapping task (see Fig. 6.5 in the text), movement time is proportional to the amplitude (distance from centre of target to centre of target) and target width (distance between the two lines defining the target location). Fitts found that average movement time increased (gets slower) as the amplitude (A) of the movement increased or the target width (W) decreased. Specifically, Fitts Law suggests that movement time is directly proportional to the ratio 2A/W. Fitts referred to this ratio, when expressed as a log2 (i.e., log to the base two), as the index of difficulty (ID). Specifically, average movement time per tap is linearly related to the index of difficulty. The index of difficulty is expressed in units of measure called “bits” (short form for binary digits).

As has likely been suggested in class by now, because motor learning/motor control is such a young scientific field compared to areas like chemistry or physics, there are few laws. Fitts’ Law is an exception. The present lab will test Fitts’ Law. Since Fitts’ Law has been shown to hold in hundreds experiments, it is hypothesized that average movement time per tap will be linearly related to the index of difficulty in the present experiment.

Methodology

Participant

One performer who is naïve to Fitts’ Law and to the intent of the lab is required (therefore not in Kin 080). He or she will participate in several different index of difficulty conditions ranging from 2 to 6 bits.

Task/Material

Three different movement amplitudes (A = 2, 4, & 8 inches) will be combined with three different target widths (W = ¼, ½, & 1 inch) to produce 9 conditions.

Ampitude Target Width Index of Difficulty

a) 2” 1” 2 bits

b) 2” ½” 3 bits

c) 2” ¼” 4 bits

d) 4” 1” 3 bits

e) 4” ½” 4 bits

f) 4” ¼” 5 bits

g) 8” 1” 4 bits

h) 8” ½” 5 bits

i) 8” ¼” 6 bits

To construct the testing sheets refer to Figure 6.5. Each condition a) through i) will be drawn on a separate sheet of paper (using the 11” dimension as the width of the sheet and the 8.5” dimension as the height). Remember that amplitude, as shown in Fig. 6.5 is from centre of target to centre of target. Condition a) is drawn below where A=2” & W=1”.

2”

1

2 X X

3

4

5

1”

If lines are drawn across the width of the sheet (11”) to make 5 approximately equal sections, then there is room on each sheet of paper for five tapping trials. The participant will perform five trials in each condition. On a single trial the participant will tap back and forth (approximately from X to X as shown above) as quickly and as accurately as possible for 10 seconds. If more than three errors are committed (i.e., the pen/pencil lands outside of the target lines) the trial must be redone. Unlike what is displayed in Figure 6.5, the participant is to begin by resting the pen/pencil down inside the target area (right handed people inside the right target and left handed participants inside the left target).

Procedure

On each trial, following the command “go” from the experimenter, the performer begins to tap back and forth as quickly and as accurately as possible until hearing the command “stop” (after 10s). If more than three errors are committed the trial is redone. If the pen hits on the target line or hits first inside the target and slides outside of the target area, these are considered hits. Only when the pen strikes outside of the target area on first contact with the paper is the tap considered an error. The experimenter will count the taps during the performance, since it is difficult to see all of the pen strikes after a trial is finished. The easiest way to count is to count each time the pen returns to the beginning position and then multiply this number by two to get total taps in 10s. Errors are checked for after each trial is over. As long as there are three or less errors on a given trial all taps count in the total number of taps for that trial.

Beginning with condition a), the performer completes two “good” trials in each condition (A good trial = a trial with 3 or less errors). These initial trials are considered practice. Following this, the performer begins back at condition a) where he or she completes three good trials. After this the performer moves on to condition b) and so on, until all conditions are completed.

Data Analysis

The dependent variable of interest is the average movement time per tap (MT) in each condition. The movement time calculations are to be completed separately for each condition a) through i). For each condition average movement time per tap is calculated as follows:

·  first calculate the average number of taps in 10s over the final three trials in each condition (e,g,, trial 3 = 23, trial 4 = 25, trial 5 = 27: average = 25),

·  to get average movement time per tap divide 10 s by the number of taps (i.e., 10/25 = .4 s (therefore each tap took, on average, 0.4 s),

·  convert this decimal number to milliseconds by multiplying by 1000 (i.e., 0.4 x 1000 = 400 ms or each tap took on average 400 ms)

·  an average MT is computed for each of the nine conditions (these nine will be plotted on a graph as described below).

Results

Construct a table that has five columns: condition (use letters above), amplitude (A), target width (W), index of difficulty (ID), & movement time per tap (MT) in ms.

Construct a figure (see below) that has average movement time per tap in ms on the Y axis and index of difficulty conditions from 2 to 6 on the X axis. There will be 9 data points on the figure.

600

MT 500

(ms) 400

300

200

2 3 4 5 6

ID (bits)

Discussion

  1. Do your results in general support Fitts’ Law? Explain (refer to figure).
  2. Would MT for the different conditions be comparable if errors were NOT controlled (i.e., performers allowed to make any number of errors/trial)? Explain.
  3. Does your constructed figure look like the one shown in Figure 6.6 in the text? Explain which figure is correct.

Writing your lab

Remember each student is to work independently. This means that students collect their own data and write their own lab report. DO NOT SHARE ANY PART OF YOUR LAB WITH ANYONE ELSE, EVEN IF YOUR FRIEND SAYS “JUST SO I CAN SEE IT”.

REMEMBER THAT YOUR WRITING SHOULD BE AS PERFECT AS YOU CAN MAKE IT. If you know you do not write well you should find an editor to check your work before handing the assignment in. In science poor grammar, unclear sentences, and poor writing style are not tolerated. While no grades are assigned directly for writing in this lab, it is unavoidable that labs that are well written will likely receive higher grades than labs written poorly, because the former will be much easier to understand. IT IS NOT THE MARKER’S JOB TO TRY TO FIGURE OUT WHAT THEY THINK THE WRITER MEANT IN POORLY WRITTEN MATERIAL.

Form for write-up

1.  Maximum word count for report is 350 words (not including Tables/Figures and Appendix ). Put rough calculations and individual tapping sheets in Appendix.

2.  On top right (as a header) put your name, student number, and word count.

3.  Your report will have the following parts:

·  title (centred at top of page)

·  2-3 sentences as introduction including your research hypothesis (what you think will happen to performance and why).

·  Results section (titled and centred “Results”) – ideally computer generated tables and figures of the results.

·  Discussion section (titled and centred “Discussion”) – number your answers to the three questions.

Marks [there will be a bonus of 20% attached to parts of this lab with the maximum possible for the lab of 100%]

1.  Overall presentation of paper, including introduction 20%

2.  Results (presentation & clarity of tables & text) 20%

3.  Question one 20%

4.  Question two 20%

5.  Question three 20%