Open Access

JSON Export

{"created":"2022-01-31T15:36:38.740370+00:00","id":"lit23169","links":{},"metadata":{"alternative":"Studies from the Yale Psychological Laboratory","contributors":[{"name":"Bliss, Charles B.","role":"author"}],"detailsRefDisplay":"Studies from the Yale Psychological Laboratory 1: 1-16","fulltext":[{"file":"p0001.txt","language":"en","ocr_en":"INVESTIGATIONS IN REACTION-TIME AND ATTENTION\nBY\nCharles B. Bliss, Pii.D.\nIntroduction.\nThe work described in the following pages occupied the greater part of my time during the academical year 1892-93. As I was the first to carry on such experiments in the Yale psychological laboratory, a large part of my fall term was spent in preparing the apparatus and in developing a method which should serve for all future experiments. The result is a method for measuring reaction-time which is in some parts entirely new. In operation it is simple and accurate, having been built up step by step as the needs required. In the hope that the whole or parts of it will be of value to other laboratories, the description has been made as complete as seemed necessary.\nIn the experimental part of the work I am especially indebted to the following persons for valuable time which they have spent in the reaction-room. To Messrs. Thomas J. Lloyd, William I. Cranford and Joshua A. Gilbert of the Graduate Department, and to Mr. Joseph Roby, a member of the senior class in Yale College.\nDuring the second term, Abraham Fisher, the laboratory steward, recorded all the experiments, thus leaving me free to do my own reacting. The advantage of doing my own introspective observing was an important one.\nDr. Scripture not only suggested the first problem but has always been ready to assist me in carrying out the experiments and in arranging the apparatus. In fact, parts of the apparatus were invented by him. One line of research was carried out at the suggestion of Professor Ladd, who has always shown a kindly interest in my work.\nIn the drawing of the diagrams valuable suggestions were received from Mi-. Walter I. Lowe, a member of the Graduate Department.\n1 \u2022","page":1},{"file":"p0002.txt","language":"en","ocr_en":"2\tInvestigations in reaction-time and attention.\nApparatus.\nApparatus for measuring reaction-time must furnish some means for giving the reactor a stimulus and for measuring the interval of time between the moment in which the stimulus is given, and that in which the reaction takes place. The time-measurement must be accurate to thousandths of a second and the person experimented upon must, so far as possible, be free from all influences which would distract his attention.\nThis last requirement was met by placing the reactor in a separate room, so constructed as to be free from light and sound. In the center of the building a room was finished off, twelve feet long, nine feet wide and nine feet high. Inside of this room a smaller one was constructed with a door and ventilator corresponding to those of the outer room. This inner room was supported on thick cushions of felt and rubber, the only connection with the outer room being heavy canvas around the doors and the ventilators for the purpose of holding back the sawdust with which the space between the two walls was filled. The door was likewise made double with beveled edges, like a safe door, so that it shut tightly against the canvas connecting the two rooms. A thick mat, made of hair felt and covered with cloth, was hung up over the door on the inside. This acted like a heavy curtain to check any sounds which might creep in around the door.\nDuring the experiments the door of the dark room was not shut more than five or ten minutes at a time. For that period the ventilator could be kept closed without producing any bad effects. In the case of longer experiments it is proposed to open a ventilator in the floor and pass a current of air through the room by means of a blower. The ventilators can then be packed with wool or some other material, which will allow the passage of air, but effectually shut out all sound. The experiments described in this paper were all taken in the winter, and the temperature of the room was the same as that of the rest of the building. Very loud sounds in adjacent rooms can still be heard in the reaction-room. Heavy wagons, which occasionally pass along the street, jar the whole building and with it this room ; the shaking can be felt but not heard. When the adjacent rooms are kept quiet, the reaction-room is free from sound. The reactor is thus practically removed from all external disturbances in sight or hearing.\nThere are two methods in use for measuring intervals of time to thousandths of a second, the graphic method and that of the chrono-","page":2},{"file":"p0003.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\n3\nscope. The Hipp chronoscope is the most perfect piece of mechanism thus far constructed for recording such short intervals of time on a dial. An immense amount of time and labor has been spent in perfecting this chronoscope and in investigating its accuracy. In its most perfect form there is always a very large error in the results as they are read off from the dial. This error depends on the relative strength of the electric current passing through it and that of a spring which pulls back the armature when the circuit is broken. A control-apparatus must be used which consists of a hammer so arranged that it can be made to fall certain distances. The time required for this fall is carefully measured by the graphic method and the spring of the chronoscope adjusted until the chronoscope itself measures the time of fall with the same result. Other times are accurately obtained by correcting the recorded results. The chronoscope in one of its forms is then accurate only for times of about that length. G. E. M\u00fcllek claimed1 that M\u00fcnsterberg\u2019s experiments contain a large error even though he had corrected his chronoscope by a control-hammer. That particular hammer was made to correct the chronoscope for intervals of 160 thousandths of a second, whereas many of M\u00fcnsterberg\u2019s experiments gave times as high as half a second. Although M\u00fcnsterberg seems to have avoided the error supposed, yet the danger is evident. A more elaborate control-hammer has been constructed by Wundt.2 By means of this hammer correct times can be given up to 616 thousandths of a second. The mean variation of this hammer in 200 experiments was 1.04 thousandths of a second. The mean variation of chronoscope and hammer combined was also 1.04ff. This the best result which has yet been obtained from the chronoscope, is ten times as great as the mean variation of the graphic method in its simple form.3\nIn the graphic method a tuning-fork, kept in constant vibration by a current of electricity, is allowed to trace a curve on a revolving drum covered with smoked paper. This gives a representation of a period of time divided according to the rapidity with which the fork vibrates. Using a fork which vibrates one hundred times a second the drum is revolved with such rapidity that the single waves are so long that we make no error in estimating tenths of a vibration and so reading the results in thousandths of a second.\n1\tG\u00f6ttingisehe gelehrte Anzeigen, 1891, p. 398.\n2\tK\u00fclpe and Kirschmann, Ein neuer Apparat zum Contr\u00f4le zeitmessender Instrumente, Phil. Stud. 1892 VU 145.\n3\tWundt, Physiol. Psych. 3 ed. II 282.","page":3},{"file":"p0004.txt","language":"en","ocr_en":"4\tInvestigations in reaction-time and attention.\nNow, given this tuning-fork curve, all that is needed is some method of registering alongside of it the exact instants at which the stimulus and the reaction occur. Fig. 1 shows the usual way in\nFig. 1.\nwhich this is done. The upper curve is drawn by the recording point of a tuning-fork which vibrates one hundred times a second. The other two lines are drawn by electro-magnetic time-markers. The current passing through one passes also through the key whose closing produces the stimulus. The current through the other passes through a key in the reaction-room. The movement of each key is thus recorded hy a break in the straight line drawn by its time-marker. These points are then transferred to the tuning-fork curve by dropping perpendiculars from the points to the curve. The measure of the time which has elapsed between the movement of the two keys can then be counted off on the time-curve.\nThe objection to the use of this method in making a large number of experiments is that it takes a long time to drop the perpendiculars and that great variable errors are likely at the two points. These errors are increased by the fact that the time-markers must be adjusted so that they shall both touch the drum in the same perpendicular line. Moreover, the latent times of the markers may not be the same.\nFig. 2.\nAn improvement upon this method has been made by doing away with one of the markers. The same current is passed through both keys and through one time-marker. Fig. 2 shows a measurement made by this method. Closing the stimulus-key draws the lever of the marker toward its magnet, making a break in the straight line. Opening the key in the reaction-room breaks the circuit and allows the lever to fly back again, thus making a second break in the same straight line. These two points are transferred to the time-curve and the interval is counted off as before. The result is that the","page":4},{"file":"p0005.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\nnumber of lines on the smoked paper is reduced by one-third, allowing more experiments to be taken on the same paper and making the records easier to count. But more important than this is the fact that one large source of error is removed. The accuracy of the result no longer depends on the adjustment of the two markers so that they shall touch the drum in the same horizontal line.\nThe latent time for the two movements is generally different. And there still remains an error and a great loss of time in transferring the two points to the time-curve. What is wanted is some means of registering the interval directly upon the curve itself. This has been accomplished after trial of various methods. The first suggestion was to arrange an apparatus so that the stimulus-key when it was closed should start the curve and the reaction-key stop it by being opened. This was done by taking the fork from the drum and replacing it by one of the electro-magnetic markers. The current was run through the tuning-fork, the time-marker and the reaction-key, but it was short-circuited through the stimulus-key.\nFig. 3.\nAs long as the stimulus-key remained untouched, the marker did not vibrate; but as soon as it was touched, the record began. When the reaction-key was pressed, the entire circuit was broken and the record ceased. Such a record is shown in tig. 3.\nAt first sight this might appear to solve the problem but a closer examination shows that this is not the case.\nFig. 4 shows the way in which the Pfeil marker works. B is the battery, F the tuning-fork, M an electro-magnet, S a steel armature which serves as a spring ; the lever is attached at (). This lever swings on a pivot at II; when the circuit is complete the vibration of the fork alternately makes and breaks the current at X. As soon as it is made the coil in the fork becomes a magnet, pulls the prongs inward and breaks the circuit. This demagnetizes the coil, the prongs fly back and the process is repeated indefinitely. But when the current is closed at X, the magnet .AI draws down the armature and its lever. When the current is broken in the fork the armature flies back carrying the lever with it. Thus the point P vibrates in unison with the fork.","page":5},{"file":"p0006.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\nLet CDE be a section of the curve which would be traced by the marker. From C to D the motion of the point comes from the spring that causes the armature to fly back. From D to E the motive force is a combination of the spring and the magnet. Now if the stimulus-key, which starts the current through the marker, is opened between C and D, no effect will be produced until the point D is reached for no current is passing through the circuit. Therefore the chances are one in two that the beginning of the movement will be too late b3r anywhere from 0 to 5\u00b0, the section of the curve from C to E being 10a with a tuning-fork which vibrates 100 times a second.\nFig. 4.\nAt the end of the interval the case is not quite so clear. If the reaction-key breaks the circuit between C and D, there will be no efEect Until the point D is reached. But, unlike the beginning, the effect will not be shown at D. For, when the marker is in motion the motive force, between D and E is a combination of the spring and the magnet. Near D the spring is stretched. The tension gradually decreases passing from a positive to a negative quantity somewhere below the middle point, while the force of the magnet gradually increases the nearer the armature approaches it. Therefore the effect of a break in the current is not shown until the magnetic component of the motive force reaches a certain sti\u2019ength in proportion to that of the spring before it is interrupted. Suppose that this takes place when three-fifths of. the distance DE have been passed over, then the chances are seven to ten that the end of the interval will be registered anywhere from 0 to 1a too late.\nIn a large number of experiments these errors at the beginning and end would partially balance each other, but their presence would still be shown by the large mean variation. The beginnings would on the average be too late and the ends too late by 2.45ff. In a small number of experiments the results are not accurate beyond hundredths of a second. By using a fork which vibrates 500 times a second, the error would be reduced to 2a and by using a fork vibrating 2000 times a second, the method would be accurate to","page":6},{"file":"p0007.txt","language":"en","ocr_en":"Investir)citions in reaction-time and attention.\nthousandths of a second. This however is impossible since the time-markers are not delicate enough to record such rapid vibrations. Even if they were, the task of counting so many wave-lengths would render the method of no practical use.\nThe next step was to arrange the apparatus so that the time-marker vibrated continually in unison with the tuning-fork, but yet\nFig. 5.\nso that closing the stimulus-key sent an additional current through the time-marker, which additional current was released by opening the key in the reaction-room. The result is shown in %\u2022 5-During the interval to be measured the vibration of the marker continues in a different line from that of the normal time-curve. Here we have the beginning and end of the interval accurately marked. By adjusting the strength of the two currents and the rapidity of the drum, this method will probably be quite successful. If so, it will be superior to any otlier method heretofore used. It is possible however only with the Pfeil marker which has a steel spring as shown in tig. 4 ; for it consists in a partial magnetization of the electro-magnet which draws the spring part way but still leaves room for it to be affected by the current passing through the tuning-fork.\nThis method was not used in the following experiments for the reason that a much better plan suggested itself. Instead of trying to change the curve to mark the beginning and end of the interval the apparatus was so arranged that closing the stimulus-key broke the primary current of a spark-coil and sent a spark from the tuning-fork to the metal drum through the smoked paper. Opening the key in the reaction-room likewise broke the same current and sent another spark through the smoked paper. Fig. 0 gives a speci-\nFig. 6.\nmen record taken by this method. Here we have both ends of the interval marked exactly, there is no time lost and no error arising from transferring the interval to the time-curve or in adjusting the markers on the drum.","page":7},{"file":"p0008.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\nThe question now arises whether the spark occurs at the exact time the primary circuit of the spark-coil is broken. In a large dynamo the spark might be delayed several seconds, but in a small coil such as is used for this purpose there is no delay sufficient to affect the result. Fig. 7 is one of a series of experiments made to test this point. The time-curve represents hundredths of a second. The straight line was drawn by a pointer attached to the lever of a key, the least motion of which broke the primary current of the spark-coil, sending a spark from the key to the drum through the smoked paper. None of these experiments show any latent time.\nFig. 7.\nEven if there were a latent time which could be detected by reading to ten-thousandths of a second, that would not affect the reliability of measurements taken by this means. For the same coil, under the same conditions, this latent time would remain constant. The spark which marks the end of the interval would be delayed just as much as the spark which marks the beginning ; the interval between the two would be exactly the same as though both sparks were in their proper places.\nPresupposing the accuracy of the' tuning-fork, this method is absolutely accurate to thousandths of a second. In counting the records one falls into the habit of counting by threes, making use of the psychological fact that three things can be kept in consciousness at the same time as easily as one. In this way a paper containing twenty-two records of simple reaction-time can be counted in four minutes. This is a much shorter time than is required to record the readings of the Hipp chronoscope and correct them for the variable error.\nThe only justification for the use of the curonoscope lies in the supposition that it saves time. The method here described is much simpler and quicker, in addition to being absolutely accurate. For simple reaction-time this method is far more accurate and rapid than any hitherto described. But for longer times, such as association-time or discrimination-time, some easier method must be devised for counting the records. It is suggested that another time-marker be placed on the drum ; this marker to be connected with a pendulum","page":8},{"file":"p0009.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\n9\nor clock work which shall mark every third or fourth wave of the tuning-fork curve. With this assistance it will be possible to count long records quickly and accurately.\nTwo drums were used in these experiments, an electric drum and a K\u00f6nig drum.\nThe electric drum consists of a brass cylinder mounted on iron brackets so as to turn on its axis. It is rotated by a small electric motor clamped to the right side of the table. Linen thread serves for a belt. The tuning-fork, or the marker is mounted on a carriage which moves along a track parallel to the drum by means of an endless screw turned by a crank at the left side of the table. One turn of the screw moves the marker on the drum one-quarter of an inch to the right or left. By regulating the strength of the current which runs the motor considerable variation can be produced in the speed of the drum, while a switch near at hand allows the motor to be stopped or started at will.\nTwo kinds of curves are produced by this drum. In fig. 8 the marker, or the fork itself, is kept vibrating on the drum all the time\nFig. 8.\nWhen a record is taken, the marker is quickly moved to the right or left by a turn of the screw. The record is made during this side-wise movement of the marker, the result being two white bands with the record on a curve passing obliquely from one to the other. The first spark comes from the stimulus-key, the second from that in the reaction-room.\nAt first it is somewhat difficult to make these two motions, the one with the right hand, the other with the left, just at the right instant, but a short practice enables one to do it with ease and accuracy. The whole difficulty, however, is removed by the multiple key to be described below. Using that, together with the electro-\nH\nFig. 9.\nmagnetic time-marker, we produce the curve shown in fig. 9. The marker remains stationary, simply drawing a white line around","page":9},{"file":"p0010.txt","language":"en","ocr_en":"10\nInvestigations in reaction-time and attention.\nthe drum. Opening the stimulus-key starts the marker vibrating and an instant later gives the stimulus which is marked by a spark on the curve. As soon as the spark from the reaction-key has been recorded, the multiple-key is released, and the marker ceases to vibrate before the drum has made a complete revolution. The marker is then moved to the right or the left by a turn of the screw and another record is taken. A similar record on an ordinaiy drum is shown in fig. 10.\nFig. 10.\nThe other drum is constructed, as can be seen in fig. 13, so that every turn of the crank moves the drum itself half an inch to the right or left. In this case the standard holding the marker or the tuning-fork remains stationary. One hand turns the drum once around while the other closes the stimulus-key. The reaction always follows before the revolution is completed. For simple reaction-time this is much the better drum of the two. The records are always the same distance apart and can be made to begin in the same horizontal line on the drum, thus making the counting of the records much easier than those of the electrical drum which are scattered over the paper and are liable to be cut in two when that is removed from the drum. For other purposes the electrical drum is to be preferred.\nIt is evident from the various functions ascribed to the stimulus-key that something more than the ordinary telegraph-key is implied. In every case it is assumed that it produces the stimulus and records it at the same instant on the drum by means of a spark. In addition to this it sometimes starts the tuning-fork curve just before the stimulus and stops it just after the reaction has taken place. The necessity for some contrivance by which such things might be done was felt at the beginning of the work and led to the invention, by Dr. Scripture, of the multiple-key.\nFigs. 11 and 12 show drawings of this key. It consists of two square bars of brass I and H rotating on small steel axles X held in place by check-screws K L passing through the upright parts C of a firm brass frame which is screwed to the wooden base J. One end of each bar is held by a spring M, Y ; the strength of M is regulated by the screw A. Besides this there are four other screws B, D, F, F, which pass through the upper lever, one of them F","page":10},{"file":"p0011.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\n11\nbeing insulated from the lever by hard rubber. The screw B rests upon a small steel plate P, insulated bv a hard rubber screw from the brass stanchion Q and connected by wire on the underside of the base with binding-post 5. None of the binding-posts are shown in the elevation because they would conceal important parts o! the key. The screws B, D, F are connected through the brass rod, steel axles and upright support with a screw that passes through the base and thence by insulated wire with binding-post 1.\nK\tB\tC\tD\tE\tF\t____C\n\t\tP\t\u00c6 PT\n\t\t\u00abw\t\th\tn\n0\t\u00c8 m a u\u00e7d g\tfp \u00c4 p\t- Ki- \u00a5\tI- >bT\t.ftU\n\t4, i\t.id)\tT Ml FI\n\tFoT\t\u20220 \n\tj\t>\tp\t\nThe screw B being adjusted so that the upper lever is level, the screw F regulates the amplitude of its movement. By means of this screw it makes contact with the brass stanchion Y. The screw E which is insulated from this lever, is connected to post 2 bv insulated wire running along the lever, down the standard through the base. The screw l with which it comes in contact is also insulated from its lever and connected in a similar way with binding-post 3. Bv means of screw 1) the copper spring S can be made to make contact with screw T which is insulated from its lever and connected with binding-post 0.\nThe lower lever is adjusted to a level position by screw O ; it is insulated from its steel axle and together with screw () is connected with binding-post 7. The screw Z, with which it makes contact, is connected with binding-post 4. So arc the brass stanchion Q and the screw N, passing through it, as well as the mercury cup W. The","page":11},{"file":"p0012.txt","language":"en","ocr_en":"12\nInvestigations in reaction-time and attention.\nscrew E is so regulated that just before F makes contact with Y the lever I breaks contact with the screw Z and immediately after makes contact again in the same circuit either through the screw N or the mercury cup W.\nWe have six contacts : three makes, two breaks, and one break followed by a make in the same current. One of these breaks, if used at all, must always come first and one of the makes, if used at all, must always come last. According to actual count this gives forty-four different ways in which currents can be passed through the key. When more than one current is being passed through the upper lever at the same time, care must be taken to have this lever connected with the same pole of all the batteries.\nA few of the uses to which this key may be put will be mentioned here together with the contacts used in each case.\n1.\tAs an ordinary key where the contact is made by pressing down the key ; circuit through E-U or F-V.\n2.\tAs an ordinary key in which the contact is broken by pressing down the key ; circuit through B-P or O-Z.\n3.\tTo close two circuits at the same time ; E-U, D-T.\n4.\tTo close to circuits at the same time and one an instant later ; D-T, E-U, F-V.\n5.\tTo close one circuit and break another at the same time ; E-U, O-Z.\n6.\tTo close two circuits and break a third at the same time ; E-U, D-T, O-Z.\n7.\tTo break one circuit just before closing a second ; B-P, E-U.\n8.\tTo break a circuit and an instant later close the same circuit again ; O-Z, R-N, or I-W.\nIn the second method of recording reaction-time on the smoked drum, according to the arrangement of wires in fig. 13, the tuning-fork current is short-circuited at P-B while the key remains closed. When this contact is broken, the current passes around through the marker on the drum. A moment later the contact E-U closes a telephone-circuit which passes through the apparatus in the reaction-room and so produces the stimulus. But the primary current of the spark-coil is passing through O-Z. At the same instant in which the contact E-V is made, this contact O-Z is broken and a spark passes through the smoked paper. This primary circuit is made again at W in time to be broken a second time by pressure upon the key in the reaction-room. As soon as this reaction takes place the operator releases the multiple-key, the tuning-fork curve is short-","page":12},{"file":"p0013.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\t13\ncircuited again at B-P and the time marker ceases to vibrate. As the reaction always follows the contact E-U within three-tenths of a second, the key need not be kept open longer than is natural in slow movement.\nFig. 13.\nThe most important pieces of the apparatus having been described in detail, its general arrangement can easily be understood from the diagram in fig. 13. The reaction-room is indicated above on the left. The room used for the production of sound stimuli is shown next to it, although it is situated in another part of the building so that","page":13},{"file":"p0014.txt","language":"en","ocr_en":"14\tInvestigations in reaction-time and attention.\nno sound from the loud tuning-forks can penetrate the walls of the reaction-room. The recording-room is on the floor below. These figures are all diagrammatic, being drawn to show the meaning of wires and apparatus rather than actual positions or proper proportions.\nTaking the diagram from left to right, the first pair of wires belongs to an electric light circuit.. The lamp 1, which was used in these experiments, was a miniature incandescent lamp of 6 c. p. By a switch, 10, the light could be turned on or off.\nThe next three wires connect two electric bells, 2 and 12, with the Leclanch\u00e9 elements B and C. Closing key 3 rings the bell in the recording-room. The gong being removed from bell 2 and the contact made permanent, closing key 11 only produces a click in the reaction-room. Otherwise the sound would be so loud as to distract the person reacting.\nThe next pair of wires forms a telephone-circuit by means of which the experimenter can talk freely with the person in the reaction-room. The switch 13 breaks this circuit during a series of experiments and so prevents any noise reaching the reaction-room through this telephone. This telephone connection is a new feature in reaction-time apparatus and its advantage cannot be overestimated. In some of the German laboratories the reactor and experimenter are in the same room, separated only by a cardboard partition. The reactor is thus influenced by every sound in the building, by the changing lights and shadows and by the noise of the chronoscope. In other laboratories the reactor is placed in a separate room in another part of the building. When the operator desires to speak to the reactor he must leave his work and go to this room, often breaking up the whole series and producing more or less distraction. With this arrangement the reactor is in a dark room, free from sound. By a turn of the switch he can hear even a faint whisper from the experimenter.\nNext on the diagram comes a one-inch Ritchie spark-coil ; 31 and 30 are the poles of the primary circuit. The current from battery E, consisting of two to four Grove cells, passes through the closed key 6 in the reaction-room and through the contact 21-22 in the multiple-key (O-Z of fig. 11). When the key is pressed down the circuit is broken at 22 and closed immediately after at 23. In practice it was found better to use the mercury cup W, fig. 11, for this second contact as the contact with the copper spring and iron screw R, is a sliding contact liable to produce additional sparks and thus to con-","page":14},{"file":"p0015.txt","language":"en","ocr_en":"Investigations in reaction-time and attention.\n15\nfuse the record. The mercury must be kept covered with water as alcohol takes fire with a current of the size used.\nThe poles of the secondary coil are 28 and 29. One was connected with the brass cylinder of the drum by being attached to the iron frame at 16, the other with the point which marks on the drum. When the electro-magnetic time-marker was used, a light aluminium point was substituted for the ordinary straw or quill point. Every time either key is pressed a spark passes from the marker to the drum through the smoked paper scattering the smoke and making the white dots shown in figs. 6-10.\nNumbers 19 and 20 are two ends of a second telephone circuit. II is the battery, 9 the transmitter and 7 the receiving telephone. Before the transmitter there stands a tuning-fork 8 run by the battery G. When the multiple-key is pressed down, this telephone circuit is closed at 19-20 and the tuning-fork is heard in the telephone in the reaction-room. At the same instant a spark is made on the drum by the breaking of contact 21-22. The strength of this sound was regulated by passing the telephone-current through a resistance-board, not shown in the diagram. For purposes of simple reaction it was not necessary to use tuning-fork 8. The short, sharp click made in the telephone by closing the circuit at 19-20 was sufficient. By changing the resistance in the telephone-circuit this sound could be varied from one too weak to be heard to one too loud to be endured. It was found necessary to run the wires of this circuit from the recording-room to the reaction-room without allowing them to come near any other wires which were in use at the same time. Otherwise sounds were produced in the telephone by induction from the currents in those wires.\nDuring the latter half of the experiments the two receiving telephones, 5 and 7, in the reaction-room were each connected with both transmitters, 4 and 9. By this arrangement the sounds from the recording-room and from the sound-room were heard in each telephone. One of them was fixed by rods and clamps in such a position, that the right ear of the person experimented upon naturally rested against it. The other was held to the left ear by the left hand, while the right hand was free to manipulate the reaction-key.\nA still better plan would be to use a head telephone with a receiver at each ear. This would always be in place, leave both hands free and allow the person reacting to take the most comfortable position and to move about instead of keeping the body in one fixed position. However, as a series of experiments never lasted over five","page":15},{"file":"p0016.txt","language":"en","ocr_en":"16\tInvestigations in reaction-time and attention.\nminutes, the disturbance from the act of holding one of the telephones cannot have been very great.\nThe remaining wires shown in the diagram all have to do with the time-curve. The tuning-fork 26 is run by a dip-battery F. This current passes through the contact 24-25 when the key is closed as shown in this figure. When the key is opened, this contact is broken and the current passes around through the electro-magnetic time-marker 17 communicating to its lever the vibration of the tuning-fork.\nIn most of the experiments made with the hand-drum this time-marker was not used. The fork itself was placed on the standard and allowed to vibrate continuously on the smoked paper.\nIn addition to the wires shown in the diagram another pair was used to connect an electro-magnet in the reaction-room with a battery and switch in the recording-room.\nAll of these wires are part of a system of wires running through the whole building. At first seven wires were laid from each room to a switch-board in a central position. This number not being sufficient for the currents required between the reaction and recording-rooms seven more were laid to each of these rooms. All of these wires where the resistance is of small importance, such as the telephone and bell circuits, are number 16 B. and S. office wire. For th\u00e9 electric light and the primary circuit of the spark-coil, where stronger currents are required, number 10 heavily insulated wire w\u2019as put in. These were run directly from the reaction-room to the recording-room independent of the switch-board.\nExperiments in reaction-time.\nIn all the experiments the stimulus to which the person in the dark room reacted was a sound produced in the telephone.\nA warning click was given on the bell in the reaction-room just before each experiment. Experiment has shown that when the interval between warning and stimulus is always the same the mind is soon able to estimate the interval correctly and always reacts just at that time whether it hears the stimulus or not. Therefore this warning cannot be produced by any mechanism connected with the drum but must be given by the voluntary act of the experimenter. The effect of this warning on the reaction-time depends on the interval between the warning and the stimulus. If the interval is too short there is not time enough to concentrate the attention and the warning hinders the reaction instead of helping it. If the time is too","page":16}],"identifier":"lit23169","issued":"1892-1893","language":"en","pages":"1-16","startpages":"1","title":"Investigations in reaction-time and attention: Introduction","type":"Journal Article","volume":"1"},"revision":0,"updated":"2022-01-31T15:36:38.740376+00:00"}