Individual Differences in Executive Processing Predict

1

Individual Differences

Individual Differences in Executive Processing Predict

Susceptibility to Interference in Verbal Working Memory

Supplementary Materials

Trey Hedden

StanfordUniversity

Carolyn Yoon

University of Michigan

Methods

Description of Tasks

Sample Characteristics

Demographics questionnaire. This self-report questionnaire collected information about the participants’ age, education level, gender, race, primary language, marital status, and occupational status.

Health questionnaire. This self-report questionnaire collected information about the participants’ health status. The health comparison item was measured on a 5-point scale in response to the question “Compared to other people your own age, how would you rate your physical health?”, where 1 corresponded to “much worse than average” and 5 corresponded to “much better than average”. The health problems item was measured on a 5-point scale in response to the question, “How often do health problems stand in the way of your doing the things you want to do?”, where 1 corresponded to “never” and 5 corresponded to “always”. For the health limitations item, participants indicated whether their health limited their ability to perform 10 categories of everyday activities (e.g., vigorous activities such as running or lifting heavy objects, moderate activities such as moving a table or pushing a vacuum cleaner, and other activities such as carrying groceries or climbing a flight of stairs). For each category, participants indicated whether they were “limited a lot”, “limited a little”, or “not limited at all”. The dependent measure was the number of categories (out of 10) that participants indicated was not limited at all. The number of medications item is the sum of the prescription and nonprescription drugs (not including vitamins or nutritional supplements) that participants use at least once a week. The health conditions item is the sum of the number of current health conditions indicated on a checklist of 18 moderate to severe conditions (e.g., asthma, chronic migraines, high blood pressure, diabetes).

Mini-MentalState Examination (MMSE). This dementia screening instrument was developed by Folstein et al. (1975) and was used to exclude participants with scores of 24 or below, which indicates possible dementia (Folstein et al, 1975; Lezak, 1995). The MMSE tests orientation and awareness, and basic cognitive and motor skills.

Executive Function: Shifting

There were three tasks designed to measure the shifting component of executive function.

Plus-minus task. This task was adapted from Jersild (1927), Spector and Biederman (1976), and Miyake, et al. (2000). The task consisted of three lists of 30 two-digit numbers. For each participant the numbers 10 to 99 were randomly assigned to the three lists. On the first list, participants were instructed to add 3 to each number and enter that response. On the second list, participants were instructed to subtract 3 from each number and enter that response. On the third list, participants were instructed to alternate between adding 3 to and subtracting 3 from each successive number. Participants had to internally keep track of whether they were to add or subtract to each number, as there were no external cues provided. Each stimulus number was presented on the computer screen until the participant entered a response using the number pad on the keyboard. The participant’s response was displayed in a response box above the stimulus number and could be edited by the participant until the answer was submitted by pressing the “Enter” key. For each list, the mean reaction time to correctly respond to each stimulus number was calculated. The dependent measure was the cost of shifting between addition and subtraction operations, calculated as the difference between the mean reaction time on the alternating list and the average of the mean reaction times on the addition and subtraction only lists. Values were calculated for each participant after trimming responses in excess of 2.5 SD from the mean of the participant’s data. Reaction times were collected with millisecond timing, and are reported in seconds.

Wisconsin Card Sorting Test (WCST). This task was a brief version of the standardized WCST and was administered using the WCST-64: Computer Version for Windows Research Edition software (Psychological Assessment Resources, Lutz, FL). In this task, participants were shown a series of single target cards at the bottom of the computer screen and four sorting cards at the top of the screen. Participants were instructed to categorize each successive target card by matching it with one of the four sorting cards. Cards could be categorized according to one of three stimulus attributes – color (red, green, blue, or yellow), number (1, 2, 3, or 4), or shape (circle, square, diamond, or star). Participants indicated which sorting card they thought each target card belonged with by pressing one of four keys, each of which corresponded to one of the four sorting cards. Following each response, participants received visual feedback on whether they had correctly sorted the target card. After 10 correct responses, the sorting criterion changed to a new stimulus attribute. Participants sorted a total of 64 cards. The main dependent measure was the number of classical perseverative errors, which was the number of times participants failed to change sorting criteria and continued to sort using the previous criterion.

Trail Making Test. This task was adapted from Spreen and Strauss (1998), and consisted of two forms. Form A consisted of 25 circles arranged on a page, with each circle numbered 1 to 25. Participants traced a line between successively numbered circles as quickly as possible. Form B consisted of 25 circles arranged on a page, with each circle either numbered 1 to 13 or lettered A to L. Participants traced a line to alternatively connect numbers and letters (i.e., 1 to A to 2 to B, and so on) as quickly as possible. A practice page was completed before the test page for each form. An experimenter recorded the time to complete each form in seconds using a stopwatch. The dependent measure was the switching cost, calculated as the difference in time to complete Form B and Form A.

Executive Function: Updating

There were three tasks designed to measure the updating component of executive function.

Letter memory. This task, adapted from Morris and Jones (1990), and Miyake, et al. (2000), involved the serial presentation of letters in lists of varying length. The task was to recall the last 4 letters in each list by writing them on an answer sheet. Twelve lists, three lists each of 5, 7, 9, and 11 letters in length, were presented. Each letter was displayed on the computer screen for 2000ms. Participants were instructed to rehearse the last 4 letters presented aloud. As each successive letter was presented, participants added the most recent letter to their rehearsal list. Once four letters were being rehearsed, participants dropped the 5th letter back from rehearsal as each new letter was added. At the end of each list, the word “RECALL” was presented to prompt participants to write down the last 4 letters in the list. The dependent measure was the number of trials in which all of the last 4 letters were correctly recalled in the correct order.

Backward digit span (BDS). This task was adapted from the WAIS-III (Wechsler, 1997a). The experimenter read aloud a series of digits to the participant, who responded by repeating the same series of digits in the reverse order (i.e., 9-1-7 for 7-1-9). The presentation rate was one digit per second. There were two trials per block, with the number of digits per trial ranging from 2 to 8. The task was discontinued when the participant incorrectly responded to both trials of a particular block. The dependent measure was the number of trials in which all of the numbers were correctly recalled in the reverse order.

Self-ordered pointing (SOP). This task was adapted from Petrides and Milner (1982), and Shimamura and Jurica (1994). The participant was presented with a series of sixteen pages containing random rearrangements of 16 abstract visual textures in a 4 x 4 array. For each page, the participant was instructed to mark off one pattern of their choice. However, as participants progressed through the pages, they could not mark any of the patterns that they previously marked. Participants completed two repetitions of the task. The dependent measure was the average number of patterns correctly marked (i.e., not marked more than once in a series) in each repetition.

Executive Function: Inhibition

There were four tasks designed to measure the inhibition component of executive function, two tasks each for prepotent response inhibition and resistance to proactive interference.

Prepotent Response Inhibition

Antisaccade. This task was adapted from Roberts, Hager, and Heron (1994), and Miyake, et al. (2000). On each trial, a fixation cross was presented in the center of the computer screen for a variable amount of time (ranging from 1500 to 3000 ms in 250-ms intervals). A visual cue (0.4º) was then presented on one side of the fixation cross for 225 ms, followed by the presentation of a target stimulus (2.0º) on the opposite side for 150 ms before being masked by a gray cross-hatched pattern. The visual cue was a black square and the target stimulus was an arrow (pointing left, right, or up) inside a box. Cues and targets were both presented 4.19 in. away from the fixation cross (on opposite sides) and participants were seated 22 in. in front of the computer screen (hence, the visual angle from cue to target was 21.6º). The participant was instructed to suppress the impulse to look at the cue, and instead to look at the side opposite from the cue and to indicate the direction of the arrow by responding with a button press. Participants practiced on 21 trials before completing 84 test trials. The dependent measure was the proportion of correct trials.

Stroop color naming. This task was adapted from Stroop (1935) and Miyake, et al. (2000). Participants were instructed to indicate (with a button press) the color of a stimulus as quickly as possible on each trial. Trials included 72 neutral trials with a string of asterisks printed in one of four colors (red, green, blue, or yellow), 60 conflict trials with a color word printed in a different color (e.g., RED printed in green), and 12 congruent trials with a color word printed in the same color (e.g., RED printed in red), with the trial types randomly intermixed during presentation. Participants completed 9 example trials before beginning the task. The dependent measure was the reaction time difference between conflict and neutral trials. Values were calculated for each participant after trimming responses in excess of 2.5 SD from the mean of the participant’s data. Reaction times were collected with millisecond timing, and are reported in seconds.

Resistance to Proactive Interference

Excluded letter fluency (ELF) 1 and 2. This task was adapted from Bryan, Luszcz, and Crawford (1997). Participants were given 60 seconds to write down as many words as possible that do not contain a given letter. Proper nouns, numbers, and variants of previously produced words using a different suffix were not allowable responses. Participants completed two versions of the task, using the letters E and A, that were separated by intervening tasks. The dependent measure for each version was the number of correctly produced words minus the number of incorrectly produced words.

Semantic fluency. This task was administered with paper and pencil. Participants were given a category name (animals) and were instructed to write down as many members of the category as possible in 60 seconds. Proper nouns and the same word with different suffixes were not counted as correct. The dependent measure was the number of correct responses minus the number of incorrect responses.

Verbal Memory

There were two tasks intended to measure verbal paired-associate memory, each of which was adapted from the WMS-III (Wechsler, 1997b).

Immediate verbal paired associates. Participants were presented with a list of 8 word pairs (e.g., OBEY – INCH) and instructed to remember the word pairs. Pairs were presented for 1.5 seconds with a 1.5 seconds interstimulus interval. After initial presentation, participants were presented with the first word in each pair (e.g., OBEY – ?) and instructed to write down the associated word. Participants were given 6 seconds to write down their responses, with a 1 second interstimulus interval. Participants received three repetitions of the same list, with a random reordering of the pairs at each repetition. The dependent measure was the average number of correct responses.

Delayed verbal paired associates. After a filled interval of approximately 20 minutes, participants were presented with the first word in each pair from the immediate verbal paired associates task and instructed to write down the associated word. Participants were given 10 seconds to provide a response, with a 1 second interstimulus interval. The dependent measure was the number of correct responses.

Visual Memory

There were two tasks intended to measure visual paired-associate memory, each of which was adapted from the WMS-III (Wechsler, 1997b).

Immediate visual paired associates. Participants were presented with a list of 6 pairings of an abstract figure with a color patch and instructed to remember the pairs. Pairs were presented for 3 seconds with a 500 ms interstimulus interval. After initial presentation, participants were presented with the abstract figures from each pair and instructed to indicate with a button press the associated color. Response buttons were labeled with the color patches. Participants responded at their own pace. Participants practiced on two unique pairs before beginning the task. Participants received three repetitions of the same list, with a random reordering of the pairs at each repetition. The dependent measure was the average number of correct responses.

Delayed visual paired associates. After a filled interval of approximately 20 minutes, participants were presented with the abstract figure each pair from the immediate visual paired associates task and instructed to indicate the associated color with a button press. Participants responded at their own pace. The dependent measure was the number of correct responses.

Perceptual Speed

Letter comparison, sections 3, 6, and 9. This task was developed by Salthouse and Babcock (1991) and was adapted for administration on the computer. Participants were presented with pairs of letter strings consisting of 3, 6, or 9 letters each. Participants determined whether the two strings were the same or different and responded by pressing a key corresponding to a “match” or “nonmatch” decision. They were given 45 seconds to complete as many items as possible at each level (3, 6, or 9 letters). For each section, the dependent measure was the number correct minus the number incorrect.

Outcome Measures

List memory. This task was adapted from the task described by Hedden and Park (2001, 2003). Participants were provided a pair of headphones with a microphone and told that their voice would be recorded during the experiment. Participants were given two blocks of 20 trials. In one block, exclusion instructions were provided, in which participants were to distinguish between studied and read word pairs. In the other block, inclusion instructions were provided, in which participants were to give the same response to studied and read pairs. Order of the instruction blocks was counter-balanced across participants. During the task, word pairs were presented on a computer screen in short lists of 3 word pairs each for 40 total trials. Each trial consisted of a study list of 3 word pairs, a read list of 3 word pairs or a rest period, and a recognition list consisting of 12 word pairs. Word pairs could be of 4 types: A-B pairs were always those pairs presented on the study list; A-C pairs consisted of the same cue word as an A-B pair joined with a new associated target word; D-E pairs were presented on read lists and consisted of paired associates that were unrelated to the A-B and A-C pairs; novel pairs were never presented on study or read lists and were unrelated to all other pairs in the study. The trials were randomly distributed across two types of interference conditions: rest and read D-E conditions. An example of a read D-E trial is presented in Figure S1. On all trials, the word “STUDY” was presented in all capital letters for 2.5 seconds to prompt participants to study the following word pairs. Participants studied the following 3 A-B word pairs for 4 seconds each with a fixation delay of 500 ms preceding each pair, during which a fixation cross was displayed. Participants were instructed to study and remember these words. The study list was followed by a blank interval of 500 ms. On rest trials, the word “REST” was displayed on the screen while participants rested for 16 seconds after the study list, before being presented with the recognition list. During this interval, participants were instructed to rest and to try to remember the words on the study list. On read D-E trials, after the presentation of the study list, the word “READ” was displayed for 2.5 seconds to prompt participants to read aloud the following word pairs. Participants were shown a read list consisting of 3 D-E word pairs displayed for 4 seconds each with a fixation delay of 500 ms preceding each pair. Participants were instructed to read these pairs aloud 3 times within the 4-second interval when prompted by a tone. A tone sounded 3 times at equal intervals within the 4 seconds: at 250 ms, 1500 ms, and 2750 ms after the appearance of the word pair. A 500 ms blank delay separated the read list from the recognition list.