AN APPARATUS FOR MEASURING REACTION TIMES WITHOUT A CHRONOSCOPE

BY DAVID WECHSLER

The Psychological Corporation, New York City

This paper describes an apparatus, originally devised to measure alertness of attention and speed of reaction, which promises to be useful in the general field of experimental psychology. The apparatus consists of two units. The first enables the experimenter automatically to present several stimuli, simultaneously, successively, or both together and in sequence; the second automatically records the subject's responses as they are made. The two units may be used together or separately. When used together they furnish a means of measuring reaction-time without the use of a chronoscope.

AUTOMATIC PRESENTATION OF MULTIPLE STIMULI

Figure I shows the construction of the first unit. A metal cylinder (CL) is connected to an electric motor (M) which moves the cylinder at a very low speed by a series of reduction gears. The cylinder is covered with an insulating sleeve (SL). From this sleeve are cut out, at varying intervals and of different dimensions, windows which expose small portions of the metal. Over the cylinder are placed a number of metal fingers, f1, ƒ2, ƒ3, which make contact with the cylinder whenever one of the windows passes beneath. The fingers complete one or another of several circuits, such as fi-S1-Mg-A or f2-S2-Mg2-A. S1, S2, S3 are excitatory mechanisms; Mg1, Mg2, Mg3 are electromagnets. Both are connected in simple series with the metal fingers. The function of each of the excitatory mechanisms is to produce a particular stimulus. The electromagnets function in connection with the second or recording unit of the apparatus.

From the arrangement of the parts and the electrical con

ticular stimulus is set off and continued so long as the contact is maintained. The duration of a stimulus will depend upon the speed at which the motor is revolving and the dimension of the window; the frequency with which it appears, upon the number of times a like window passes under its corresponding metal finger. All these factors may, of course, be determined in advance by the experimenter, who can thus control the frequency and the order of his stimuli as well as their duration. Moreover, the number of different kinds of stimulus that can be employed simultaneously is only limited by the width of the cylinder.

RECORDING THE RESPONSES

The mechanism of the second unit of the apparatus (Fig. 2) enables the experimenter to obtain an automatic record of the reactions to a series of stimuli. Connected in series with the source of current (25 volts) are several make-and-break mechanisms (R1, R2, R3), e.g., push buttons or telegraphers keys, certain of which the subject uses in response to the various stimuli, and likewise a number of electric counters

(C1, C2, C3) which will register these responses when the individual circuits of which they form part are entirely closed or completed. But each of these circuits can be completed only if already partially closed at the points P1, P2, P3, which are controlled by the electromagnets Mg1, Mg2, Mg, in series with the stimulus-provoking mechanisms shown in Fig. 1. Thus, when the stimulus S1 is presented, the current flowing through the circuit fi-S1-Mg1-A (Fig. 1) causes the armature, a1, to be attracted to the metal block, b1, thereby closing the circuit a-b-p-R-B (Fig. 2) at point p1. But the circuit is still

open at point R1 and can be completed only when the subject makes contact at this point by reacting with the appropriate response. Hence, if R1 is the correct response to S1, the circuit can be completed only when the subject reacts with R1. If he accomplishes this within the time that the stimulus is being maintained, the response will be recorded by the counter. C1; if he responds too late, or with any other reaction, such as R2 or R3, the circuit cannot be completed and the counter C1 will fail to register. Nor will any of the other counters (e.g., C2 or C1) register, since the circuits of which they form part are open at the points p2 and p3 respectively.

The automatical record of responses to various forms of stimulus is made possible, as indicated above, by the use of elec

tric counters which both record and total the subject's reactions as they are made. The only task left to the experimenter is to note the counter-readings at the beginning and at the end of the experiment. The difference between the two figures is the number of correct reactions made within the permitted time. Errors or wrong reactions could also be recorded by the use of additional counters and by certain modifications in the electric circuits. For the sake of simplicity these have not been considered.

MEASURING REACTION TIME WITHOUT A CHRONOSCOPE

The apparatus described above offers a very easy method of calculating reaction-times without the use of a timerecording instrument. This is made possible by the fact that it supplies a means of automatical control of the duration of the stimulus and of the time within which the correct reaction may be made. The duration of the stimulus is determined by the speed of the cylinder and the size of the windows cut from the insulating sleeve. Given the speed the windows are made of such dimension that the duration of the stimulus will be of any desired length. A sufficiently large number of windows of varying dimensions will then furnish a graded 1 series of stimuli. The rest is a matter of making the proper connections between Unit 1 and Unit 2 of the apparatus. All stimuli of a given duration are connected with a particular electric counter which then registers the total number of correct reactions for this time; that is to say, the total number of successes for a particular reaction-time interval. From the figures thus obtained it is then easy to calculate a subject's mean reaction-time in a manner similar to that employed in calculating a limen by the method of right-and-wrong cases.2

An illustration will make the method clear. Suppose that we desire to calculate the motor reaction-time to a visual stimulus. We decide first on the limits of our durations.

1 The durations of the stimuli, should, when possible, vary by some constant interval. This simplifies the calculations.

2 In our experiment a successful reaction would correspond to a correct (right) judgment, a failure to react within the time interval to an incorrect (wrong) judgment. No distinction is made between a failure to react and an incorrect reaction.

For practical purposes 100 to 300 sigma are convenient limits. We next select the constant interval by which our reaction times are to differ. Suppose that it is 25 sigma. Then we cut windows in the insulating sleeve of such dimensions that the metal fingers will form contact with the cylinder for periods of 100, 125 ... 300 sigma. A sufficiently large number of windows are cut out to present each stimulus at least 10 times during the course of the experiment. This furnishes us with a graded series of stimuli of durations differing by a constant interval. Our apparatus enables us to record separately the successful reactions to each of these stimuli. This is done by the electric counters. All that we require at the end of the experiment is, then, to read off the totals from each of the counters and to tabulate them in an ascending or descending series thus:

It is then a matter of simple arithmetic to calculate the mean reaction-time or other desired value by means of available psychophysical formulas.

3 Allowance must be made for the 'lags' in the various parts of the apparatus, and the dimensions of the windows corrected accordingly.

The figures are hypothetical.