Title: Test of Visual Search
Technical Field
The present invention relates to a computerized test which is designed to assess the test subject's visual search speed and visual search accuracy. The ability to carry out a rapid visual survey, and to assess what is seen in that survey both rapidly and accurately, is important in a number of fields e.g., monitoring and controlling equipment, surveillance work of any kind, and driving. The test the subject of the present invention has been developed with especial reference to detecting potential problems in drivers, and therefore will be described with particular reference to this application. However, it must be appreciated that the test in fact can be much more widely applied.
In computerized tests, the subject is required to react in one or more ways to symbols appearing on a computer screen; the subject may be required to make a verbal response and/or take one or more physical actions e.g., turning a steering wheel or depressing an accelerator or brake pedal; any such physical controls are electrically connected to the computer so that appropriate movement of the controls affects the material shown on the computer screen. The test results may be recorded by the person administering the test or may be automatically recorded by the computer.
Background Art
A variety of computerized tests is available, to test for a range of different problems. For example, UK patent 2,335,856, US patents 5,344,324 and 5,103408 all relate to tests for checking that a person about to undertake difficult or dangerous work is not impaired in any way compared to their previous performance. Such tests are specific to the test subject, as the objective is to check that that particular subject is not impaired relative to their previous level of performance, rather than testing whether the subject reaches a specified general standard.
US patent 4,383,827 relates to a test which requires a complete driving simulator, with all of the car controls available to the test subject and a programme which generates a
simulation of a full road picture. The test of this type can be useful in assessing general driving ability, but is not really designed to pinpoint specific problems.
US patent 6,364,486 relates to a test of visual capabilities which requires the subject to respond to various stimuli displayed on screen either directly to the test administrator or by using a keyboard, joystick or touch screen. This type of subject reaction fails to mimic the reaction required from the subject in reality, and therefore may inhibit the subject.
US patent 5,131 ,848 discloses a test in which images are projected onto the screen in a random sequence, and the subject is required to react in a predetermined manner only to some of the images displayed; some of the test screens shown do not include any image requiring a reaction. The required reaction by the subject depends upon the type of images displayed, but the required reaction is always an instinctive reaction.
Disclosure of the Invention
It is an object of the present invention to provide a test of the subject's visual search speed and visual search accuracy which is capable of testing the subject under a range of different conditions and which requires the subject to switch randomly between instinctive and non-instinctive responses to the stimuli displayed.
The present invention provides a computerised test for visual search speed and accuracy, said test including: showing a test subject a series of different test screens each of which contains a plurality of stimuli which include a target stimulus to which the subject is required to react in a specified manner; and recording the subject's reactions; wherein:-
(a) the position of the target stimulus relative to the other stimuli is varied at random in the different test screens;
(b) each test screen is shown to the subject for a specified period; (c) the specified reaction required from the subject may be instinctive or non- instinctive.
Preferably, the specified manner of reaction involves movement of physical equipment
(e.g., for a test of a vehicle driver's visual search speed and accuracy, turning a steering wheel or using a direction indicator) and a centring screen requiring the subject to "centre" the physical equipment in the sense of moving the equipment back to a neutral position, is shown to the subject between consecutive test screens.
The target stimulus may be any of a wide range of different stimuli e.g., turn right/turn left arrows, speed limit signs, stop/go signs; or any other international road sign/signal.
The physical equipment used is selected to correspond to the target stimulus:- in the case of turn right/turn left arrows, the physical equipment used would be a steering wheel or direction indicators; in the case of speed limit signs or stop/go signs, the physical equipment would be a speedometer plus accelerator and brake pedals.
Brief Description of the Drawings
By way of example only, a preferred embodiment of the present invention is described in detail, with reference to the accompanying drawings in which:- Figure 1 shows the diagrammatic layout of the apparatus used;
Figure 2 shows the centring screen used between the multiple stimuli screens of figure 1; and
Figure 3 shows a sample screen containing a plurality of visual stimuli; Figure 4 is a flow chart showing the sequence of a typical test.
Best Mode for Carrying out the Invention
Referring to Figure 1 , the apparatus used for the test of the present invention consists of a computer 2 with a screen 2a; the computer 2 is electrically connected by a cable 3 to a steering wheel 4 which is supported in front of the test subject by suitable supports (not shown). The test subject sits on a chair 5 behind the steering wheel and facing the screen 2a. The computer 2 and screen 2a may be separated; the tester administering the test may have a separate screen which also shows the test, for supervisory purposes.
The connection between the steering wheel 4 and the computer 2 is such that the
rotational position of the steering wheel 4 is automatically recorded by the computer. The computer also is programmed to record the time at which each test screen is shown, the delay between that time and the test subject's reaction, and the nature of that reaction.
Initially, the test subject sits on the chair 5 holding the steering wheel 4 in the normal driving position and is shown a centring screen on the computer screen 2a. The centring screen 6 is shown in Figure 2:- it consists of a screen with a central box 7 and a movable arrow 8. The arrow 8 can be moved by rotating the steering wheel in the appropriate direction and the object is to centre the steering wheel into a central (i.e., neutral) position by moving the arrow 8 to the position shown in broken lines in Figure 2. Once the computer 2 records that the steering wheel 4 has been centred for a predetermined period (typically 1-2 seconds), the subject is shown the first test screen.
A typical test screen 10 is shown in Figure 3. The test screen contains a large number of stimuli (typically 60 to 70) but only one of the target stimuli (in this case a turn right arrow
11). A separate box 12 at the top of the test screen 10 shows the target stimuli to which the test subject is required to react:- in this case a turn left arrow and a turn right arrow.
When either of these stimuli appears on the screen, then the test subject is required to react by turning the steering wheel 4 in the appropriate direction i.e. left for the turn left arrow/ right for the turn right arrow.
The test subject's reaction may be instinctive or non-instinctive:- an instinctive reaction from the test subject is where the turn left arrow appears where it would frequently appear in real life i.e., on the left-hand side of the subject, or where the turn right arrow appears on the right-hand side of the screen. In the test screen 10, a non-instinctive reaction is required from the test subject:- the turn right arrow appears on the left-hand side of the screen, but nevertheless the test subject is required to turn right.
The test screen 10 is shown to the subject for a specified period, typically about 10 seconds. The subject is required to find and then react to the target stimuli as quickly as possible; the elapsed time between the first showing of the test screen and the subject's reaction is recorded by the computer 2, as is the nature of the reaction. For each test, three possible reactions may be recorded by the computer:-
a) correct reaction (where the steering wheel is turned in the correct direction); b) incorrect reaction (where the steering wheel is turned in the wrong direction); c) no reaction (where the subject fails to react at all).
As soon as the subject has moved the steering wheel 4 in either direction by more than 20°, the test screen disappears and is replaced by the centring screen of Figure 1 ; the subject then re- centres the steering wheel as described above. As soon as the steering wheel has been re-centred the centring screen disappears and the next test screen is displayed.
The target stimuli remain the same from one screen to the next, but their position on the screen is varied, so that the subject has to search for the target stimulus each time a new screen is displayed.
The above described sequence is repeated until a large number (typically 20) of different test screens have been displayed to the subject.
Before the start of the test, the test subject is given a practice run so that he or she is certain what is required.
As a result of the test, the criteria listed below are measured/calculated:-
Measures of performance
Mean response time
The mean length of time from presentation of the screen until the subject turns the steering wheel in either direction by more than 20°.
Correct left turns
The number of times the steering wheel is turned towards the left following presentation of 'turn left arrow'.
Correct right turns The number of times the steering wheel is turned towards the right following presentation of 'turn right arrow'.
Total correct turns
The total number of correct steering wheel turns. No responses
The number of times the subject does not make a steering wheel response within 10.0 s. Right/left bias
The difference in response time between right-side and left-side peripherally presented target stimuli.
Central bias
The difference in response time between central and peripheral presentation of target stimuli.
Table 1 below gives an example of typical test results
Table 1 Sample results for the Test of Visual Search
The summary of the results obtained from the test is set out in table 2:
Table 2 Example summary for the Test of Visual Search
Visual Search Test
Mean response time 5445 ms
Correct left turns 8/10
Correct right turns 6/10
Total correct turns 14/20
No responses 6
Mean right-side response time 3800ms (4/4)
Mean left-side response time 7167 ms (2/4)
Right/left bias 0.53
Mean central response time 4921 ms (4/8)
Mean peripheral response time 3990 ms (10/12)
Central bias 1.23
The subject's test performance is interpreted by considering each of the measures of performance in turn:-
1. Mean Response Time. This gives an indication of the subject's speed and efficiency of visual search; all responses are included in the calculation of mean response time, whether correct or incorrect.
2. Correct Left and Right Turns
The overall number of correct responses gives an indication of how well the subject is able to locate the target stimulus and react appropriately, or whether they are guessing.
A close to 50/50 correct response rate for both left and right steering wheel turn responses indicates that the subject is probably guessing. A high number of correct responses in one direction in association with a low number of correct responses in the opposite direction may indicate that the subject has a bias toward making a particular response, and they may be guessing.
A combination of a relatively long mean visual search and location time with high correct steering wheel turns may indicate that the subject is emphasising accuracy at the expense of speed. A combination of a relatively short mean visual search and location time with lower correct steering wheel responses may indicate that the subject is emphasising speed at the expense of accuracy.
3. No responses
A high number of no responses indicates a visual scanning and search deficit.
4. Right/left bias
A bias toward right-sided peripheral target presentations (i.e., shorter reaction times during trials when symbols are presented on the right side of the screen), may indicate neglect of the left hemispace or left visual field deficits, causing slowness in responding to target presented on the peripheral left side of the screen. A bias toward left-sided peripheral target presentations may indicate right visual field deficits, causing slowness in responding to targets presented on the peripheral right side of the screen. 5. Central bias
This measure demonstrates any larger than average discrepancies between the mean response time for targets presented near the centre of the screen, versus those presented on the peripheral edges of the screen. A larger than normal discrepancy may indicate a lack of efficiency in visual search, which could be due to visual field deficits.
The subject's performance is compared to the performance of control subjects, to gauge whether the test subject has problems in one or more of the areas tested.
It should be noted that, because the present test requires some instinctive and some non- instinctive responses, the test involves the subject's higher cognitive factors in the visual search:- the subject cannot simply assume that if he or she sees a target stimulus one side of the screen, then the wheel must be turned to that side; instead, the subject has to find the target stimulus, recognise what it is, and then take appropriate action. Because the test is conducted under time pressure, subjects with cognitive impairments are more likely to make incorrect instinctive responses to the non-instinctively positioned target stimuli. Errors of this kind draw attention to cognitive defects. Further, a subject's inability
to switch quickly between instinctive and non-instinctive responses may be a serious potential problem in on the road driving situations where this skill often is important.