TEDS Data Dictionary

Test descriptions

In the TEDS CATSLife web study battery, the four TMB tests were delivered after consent, and before Spatial Spy, in the following order:

1. Vocabulary test
2. Remembering Word Pairs test: learning phase
3. Digit Symbol Matching test
4. Remembering Word Pairs test: testing phase
5. Remembering Numbers test

The Remembering Word Pairs test was therefore effectively split into two parts, with the Digit Symbol Matching test acting partly as a "delay activity" between the learning and testing phases of Remembering Word Pairs.

Although there was no explicit logout mechanism within TMB, it was possible effectively to log out during testing by closing the browser. On logging in again, a participant would restart the current activity, if not completed. For this purpose, Remembering Word Pairs (learning and testing phases) and Digit Symbol Matching were treated as a single activity; logging out after Digit Symbol Matching and before the end of Remembering Word Pairs, then logging and continuing, in fact resulted in a repeat of the Digit Symbol Matching test (any such repeats were removed from the dataset). In other tests, partial progress through a test was erased when the browser was closed, so each test could be completed only once.

The tests could be attempted on a range of device types. On desktop and laptop computers, the response inputs were a combination of mouse/trackpad clicks and keyboard inputs. On mobile devices, the tests adapted such that all response inputs were from taps on the touch screen.

Each of the four tests is described under its heading below, with a brief outline of the testing process and a link to a detailed specification document.

Each test was preceded by full instructions, examples and at least one practice item - these are all described in the linked documents.

Vocabulary

This was a straightforward multiple-choice test of vocabulary, similar to vocabulary tests used in TEDS at early ages. There were 20 items. In each item, the participant was presented with a stimulus word, and with 5 clickable response options; the participant was asked to select the response option that they thought best matched the meaning of the stimulus word.

For each item, participants were given 10 seconds in which to respond. If 10 seconds expired without a response, a message would appear to remind the participant of the time limit, and with a button to restart the item. Hence, an item was repeated and not forfeited after such a timeout. Any given item could be timed out and repeated an indefinite number of times in this way.

Of the 5 possible responses to each item, one was judged to be correct (scoring 1) and the other four were judged to be incorrect (scoring 0).

There were no branching or discontinue rules, and every participant was presented with the same 20 items in the same order.

The vocabulary test specification (pdf) contains further details including screen shots and example items.

Digit Symbol Matching

This was a test of processing speed. Example items, with screen shots, are shown in the digit symbol matching specification document (pdf).

Briefly, every item displayed the same set of six symbols, in two rows of three, above the numeric digits 1, 2 and 3. In each item, one of the six symbols would additionally be displayed, in a larger size, above these rows of symbols. The participant was asked to recognise which digit (1, 2 or 3) this larger symbol was matched with in the display. To respond, the participant would press 1, 2 or 3 on their device's keyboard.

The aim of the test was to answer as many items (correctly) as possible within a 90 second time limit. Participants were told this in the instructions and were asked to respond as quickly as they could. There was no time limit on any individual item. There was no fixed limit on the number of items that could be displayed within 90 seconds.

Remembering Word Pairs

This was a test of memory. The test procedure is described in detail in the remembering word pairs specification document (pdf). This includes screen shots and example items.

As indicated above, this test took part in two main phases. In the learning phase, pairs of words were displayed on screen, and participants were asked to try to memorise which words went together. The pairs of words were not logically matched in terms of meaning. 25 pairs of words were shown, one pair at a time in sequence. Each word pair was displayed on screen for exactly 3 seconds, with an interval of 0.5 second between successive displays of pairs.

The learning phase was followed by the Digit Symbol Matching test (see above), which effectively acted as a "delay activity".

The second, testing phase then followed. In the testing phase, one word from each pair was displayed on screen as the main stimulus, together with four other words as clickable response options. One of the four was the correct paired word (as previously displayed in one of the pairs during the learning phase), scoring 1 if selected, while the other three words were incorrect, scoring 0 if selected.

During the testing phase, as in the Vocabulary test (above), participants were given 10 seconds to respond to each item. If 10 seconds expired without a response, a message would appear to remind the participant of the time limit, and with a button to restart the item. Hence, an item was repeated and not forfeited after such a timeout. Any given item could be timed out and repeated an indefinite number of times in this way. This mechanism only applied in the testing phase, not in the learning phase.

There were no branching or discontinue rules in this test. Every participant was presented with the same 25 word pairs in the learning phase, in the same order. Each participant was presented with the same 25 questions in the testing phase, in the same order. The order in which the word pairs were presented in the learning phase did not match the order in which the same words appeared in the testing phase.

Remembering Numbers

This was a test of memory. The test procedure, including screen shots and example items, is described in detail in the remembering numbers specification document (pdf).

In each item of this test, participants were presented with a sequence of numeric digits (1 to 9) appearing on screen. Each digit was presented for exactly 1 second. Participants were asked to memorise the sequence they had seen.

Then, participants were asked to key in the sequence of digits in reverse order. For example, having been presented with the digit sequence 1, 2, 3, 4, the correct response would be to enter the digit sequence 4, 3, 2, 1 from the keyboard.

Digit sequences increased in length from 2 digits up to 11 digits. At each sequence length, there were two items (two with length 2, two with length 3, and so on). Hence, there were a total of 20 items in the test.

The test incorporated a discontinue rule. The test would discontinue after any pair of items (at a given sequence length), if both items had been answered incorrectly. Under this rule, the number of items presented to any participant could vary between 2 and 20 (even numbers only).

The raw test score varied from 1 to 11, indicating the longest sequence of numbers that a participant had correctly remembered. The default score of 1 was reserved for any participant who had not remembered either of the shortest sequences of 2 digits (and who would then have discontinued the test).

There was no limit limit for responding to the items.

Subject to the discontinue rule, every participant was presented with the same sequences of digits, in the same order.

Test Rules

The main test rules are summarised in this table.

Dataset variables

This section briefly describes the TMB dataset variables, including both item and derived variables. See also the derived variables page for derivation details.

The 26 Year CATSLife Study TestMyBrain variables are all named with prefix zct (z=age 26, c=CATSLife, t=TestMyBrain). This prefix is followed by the test prefix (vc=Vocabulary, sn=Digit Symbol Matching, mw=Remembering Word Pairs, or mn=Remembering Numbers) and the item number. The final part of the variable name indicates the type of data, as shown below. The dataset is double entered in the conventional TEDS way, so all variable names end in "1" or "2" to indicate twin and cotwin respectively - this suffix is omitted in the summary below.

The table below summarises the TMB item variables, flag variables, scores and durations. In all item variables, XX indicates the item number (01 to 20 for Vocabulary and for Remembering Numbers, 01 to 25 for Remembering Word Pairs).

Note that detailed item variables have not been retained for the Digit Symbol Matching test, because the number of items presented was effectively unlimited. Instead, there are additional summary variables for this test, as shown in the table.

There are a few overall TestMyBrain variables, plus a few overall CATSLife battery variables, as follows.