The assessment modules in LADS Plus

Nonverbal reasoning

Nonverbal reasoning is an adaptive test involving matrix puzzles that can be solved by a careful application of logical reasoning, using both visual and verbal strategies. Each item comprises a 3 × 3 matrix with the bottom right hand square empty. The task is to choose which of six squares at the bottom of the screen complete the pattern, and then click on the ‘OK’ button to move to the next item (see Figure 1). Progress through the test depends on the person’s performance and the test is discontinued when a certain number of items within a given level are failed.

Figure 1. Example screen from the LADS Plus Nonverbal Reasoning test with Timer bar active.

The nonverbal reasoning module is not intended to be a speeded test (i.e. performed against the clock), but in the interests of avoiding excessively lengthy assessment sessions, a (fairly generous) time limit of 30 seconds has been allowed for each item. For most persons this should allow sufficient time for a reasonable attempt at each item. To allow greater time would not increase validity or reliability of the test, so if time runs out then this must be accepted as part of the exigencies of the task. The passage of time is shown by means of a red Timer bar across the top of the screen so that users can easily determine when time is running out and they must therefore come to a swift decision. However, a few individuals find this Timer bar unsettling and it may distract them from the task in hand. If this is the case, the Timer bar may be deactivated by clicking on the clock button shown in the top left-hand corner of the screen (see Figure 2).

Figure 2. Example screen from the LADS Plus Nonverbal Reasoning test with Timer bar deactivated.

Verbal reasoning

The verbal reasoning module is an adaptive test of verbal conceptual relationships. In each item two pictures are presented on the screen, separated by six words (see Figure 3). The task is to identify the word that provides the best conceptual link between the two pictures: this is the target word; the other five are distractors. For example, in Figure 3 the pictures are of a bottle of milk and a piece of cheese. Out of the six words on the list, the best word that links these pictures conceptually is ‘dairy’. Of the five distractors, two have strong links only with one picture (in this example ‘sandwich’ and ‘cheddar’ have strong links with cheese but less so with milk) and two have strong links only with other picture (in this example ‘drink’ and ‘chocolate’ have strong links with milk but less so with cheese). The fifth distractor (in this case ‘elbow’) is randomly selected. (Arguably, ‘cheddar’ has links with both cheese and milk, but the task is to find the ‘best’ link, which in this case is ‘dairy’). If the person taking the test wishes, the computer will speak the words when they are clicked on, so reading competence is not necessary. The tests phase is preceded by two interactive practice items with audio feedback on responses. Progress through the test depends on the person’s performance and the test automatically terminates when the person’s ability level has been exceeded.

Figure 3. Example Screen for LADS Plus Verbal Reasoning module.

Word recognition

The Word Recognition module is a test of lexical decoding involving speeded recognition of real word from nonwords. Five words appear on the screen in random positions (see Figure 4). Only one of these five is a real word; the other four are nonwords or misspellings of real words. The person taking the test has to click on the real word as quickly as they can. If no response is made within 30 seconds, the program automatically moves on to the next item, in accordance with the adaptive fractionation algorithm. However, for individuals who score within the top 10% of the population on the nonverbal reasoning test (classified ‘High’ Nonverbal Reasoning ability), the time on this test allowed is reduced to 8 seconds per item. The purpose of this is to place additional processing speed constraints on exceptionally bright individuals who will normally be able to compensate well for any dyslexic difficulties. In the validation studies it was found that this time restriction is still sufficient to allow all bright non-dyslexic individuals to cope with the items satisfactorily.

The test begins with four practice items, which are accompanied by spoken instructions. When the program has delivered sufficient test items to be able to make a reliable classification of the individual into one of the nine categories, the test is terminated. The minimum number of items administered is 10, and the maximum 40, although most people taking the test receive 20 items.

The cognitive processes underpinning this task are (a) rapid retrieval of real words from the mental lexicon (lexical access), and (b) swift and efficient phonological decoding to eliminate nonword distractors. Skilled readers probably carry out these processes simultaneously while quickly scanning all five words. Less skilled readers may need to process each word in succession, both phonologically and lexically. Dyslexic readers are likely to have insufficient phonological or lexical skills to cope with the task and so may have to resort to guessing for much of the time.

Figure 4. Example screen from the LADS Plus Word Recognition test

Word construction

The Word Construction module is a test of lexical encoding of nonwords from syllables. The computer speaks a three-syllable nonword (e.g. ‘Subromast’) and the person taking the test has to click on the syllables that make up this nonword in the correct order, selecting them from a nine different syllables displayed on the screen in a 3 × 3 grid (see Figure 5). As each syllable is clicked on, it appears in a template at the top of the screen. If a mistake is made, the person can click on the template and it will undo the error, so permitting another choice can be made. When the person is satisfied with their choice of syllables, they must click on an arrow at the bottom right-hand corner of the screen, and the program proceeds to the next item, in accordance with the adaptive fractionation algorithm. The test begins with two practice items, which are accompanied by spoken instructions. The task has to be completed as swiftly as possible. If no response is made within 30 seconds, the program automatically moves on to the next item. However, for individuals who score within the top 10% of the population on the nonverbal reasoning test (classified ‘High’ Nonverbal Reasoning ability), the time on this test allowed is reduced to 6 seconds per item. The purpose of this is to place additional processing speed constraints on exceptionally bright individuals who will normally be able to compensate well for any dyslexic difficulties. In the validation studies it was found that this time restriction is still sufficient to allow all bright non-dyslexic individuals to cope with the items satisfactorily. When the program has delivered sufficient items to be able to make a reliable classification of the individual into one of the nine categories, the test is terminated. The minimum number of items administered is 10, and the maximum 40, although most people taking the test receive 20 items.

Figure 5. Example screen for the LADS Plus Word Construction Test.

The cognitive processes underpinning this task are (at the very least): (a) good phonological awareness whereby the spoken word can be segmented into its constituent syllables, (b) a reliable auditory short-term working memory for holding the results of this segmentation in the correct sequence in the phonological loop while these are actively processed, and (c) an efficient system of phonological encoding whereby graphemic equivalents of phonemic codes can be recognised and assembled in the correct order. As with the word recognition task, few adults with dyslexia are likely to have phonological or working memory processes that are efficient enough for them to be able to carry out this task well. A particularly heavy load is placed on working memory because of the requirement for simultaneous processing of syllables in the grid (in order to be able to select the correct ones) whilst at the same time retaining the nonword heard in the phonological loop.

Working memory

The Working Memory module is a test of backwards digit span. A sequence of digits is spoken by the computer, and the person has immediately to enter these in reverse order from memory using the keyboard. The test begins with two practice items accompanied by verbal instructions. The test then proceeds as in a conventionally delivered digit span task, commencing with items of two digits in sequence, followed by items of three digits, and so on up to nine digits in sequence. At each level two items are presented. If correct responses are made to one or both of these items then the program proceeds to the next level, in which there will be one more digit than the previous level. If both items are incorrect, then the program terminates.12 The task has to be completed as swiftly as possible. The program allows a limited time for each item; this is a function of the number of digits in the item and varies from 14 seconds up to a maximum of 28 seconds. Figure 6 shows an example screen from the Working Memory module.

Figure 6. Example screen from the LADS Plus Working Memory Test.

The computer scores (a) the number of items correct, and (b) the number of digits in correct position. The overall score for the test is a composite of these two measures. This method of scoring provides greater sensitivity than a conventional digit span test, in which only the first method is usually employed.

A backward digit span task places a heavy load on active rehearsal processes in shortterm working memory, for which there is ample evidence of weakness in dyslexia. By contrast, a forwards digit span task arguably requires only straightforward recall from the phonological loop in short-term memory, without necessarily impinging on working memory processes. For this reason, backwards digit span is generally regarded as a more sensitive indicator of dyslexia (see Turner, 1997).

12 Except at the first level, in which the person is automatically permitted to proceed to the second level. This is a precaution against premature termination of the test in the event of the person making careless errors due to not settling into the test right away.