Interference Resolution and Aging

1

INTERFERENCE RESOLUTION AND AGING

Supplemental Materials

Cognitive Declines in Healthy Aging: Evidence From Multiple Aspects of Interference Resolution

by C. PettigrewR. C.Martin, 2014, Psychology and Aging

Resistance to Proactive Interference Tasks

Recent Negatives Task

In the recent negatives probe task (e.g., Monsell, 1978), participants heard a list of three words followed by a probe word and indicated whether the probe word was in the previous list by pressing “yes” or “no” buttons on the PsyScope button box. This task contained three trial types. On positive trials (50% of trials), the probe word was presented in the most recently presented list (list n), requiring a “yes” response. On recent negative trials (25% of trials), the probe wordwas not presented in the most recent list (list n), but it was presented in the previous trial (list n-1), requiring a “no” response. On non-recent negative trials (25% of trials), the probe word was not presented in any of the most recent lists (i.e., if previously presented, the probe was presented at least n-5 lists prior), requiring a “no” response. A 1000-ms inter-stimulus interval (ISI) separated list items, and a 2000-ms ISI separated the final list item from the probe word. Participants first received a single practice block with 10 trials, followed by a single experimental block with 96 trials. Half of these trials were positive (yes) trials and half were negative (no) trials; additionally, of the negative trials, half were recent and half non-recent negative trials. The dependent variableswereRT and accuracy for recent vs. non-recent negative trials, demonstrating a participant’s susceptibility to interference from previously relevant information.

Cued Recall Task

In the cued recall task (Tolan & Tehan, 1999; similar to Friedman & Miyake, 2004), a trial consisted of one or two lists of four words. After a filler task, participants saw a category cue and were asked to recall the item from the category that was in the most recently presented list. Approximately one third of the trials (14/40, 35% of trials) consisted of one 4-item list (one-block trials); the remaining 26 trials consisted of two 4-item lists (two-block trials). Of these two-block trials, half were “control” trials in which only the second list contained an item from the cued category. The other half of the two-block trials were “lure” trials in which both lists contained an item that matched the cued category (though participants only had to recall the item from the second list). In order to induce more interference, the list one lure was always a more frequent category member (based on strength of category membership) than the list two target. Blocks commenced with the word “Ready!,” presented for 2000 ms. Each list was preceded by a 1000-ms instruction indicating how the upcoming list should be read, either “ALOUD” or “SILENT”, and list itemswere individually presented for 2000 ms. At the start of a trial, participants did not know whether there would be one or two lists, and prior to recall, participants completed an eight second filler task. This filler task consisted of eight numbers, presented for 1 second each, to which participants made verbal magnitude judgments (greater or less than 50). Thus, one-block trials consisted of the “Ready!” signal, followed by a reading instruction (one-block trials were always read ALOUD), four sequentially presented words, the filler task, and a category cue. Two-block trials consisted of the “Ready!” signal, followed by a reading instruction (the first list was always ALOUD), four sequentially presented words, a second reading instruction (the second list was always SILENT), four sequentially presented words, the filler task, and a category cue. The dependent variable was accuracy for one list vs. two list control trials (similar to Friedman & Miyake, 2004). We also calculated the proportion of list one lures recalled in two-block lure trials.

Release From Proactive Interference Task

In the release from PI task (a variant of the task used by Peterson & Peterson, 1959; similar to Friedman & Miyake, 2004), participants completed ten blocks. In each block, participants read aloud four lists of eight items. The first three lists were from the same semantic category (using Battig & Montague’s 1969category norms) and were used to build up interference; the fourth list consisted of items from a different semantic category and was used as a release from PI trial. Lists were constructed with words ≤10letters in length, and mean category frequency was matched across lists within the same block. The words “!!!Get Ready!!!” cued participants to the start of a trial. Following this, each list item was presented individually for 1750 ms, with a 250-ms ISI between words. After the final (eighth) item of each list, participants completed a sixteen second filler task that involved counting forward by letter and number, starting with the visually presented letter–number pair (e.g., if they saw H-39, they would say “H-39, I-40, J-41…”). When cued with a green box and the word “Recall,” participants stopped counting and had 20 s for free recall. An experimenter recorded responses, including correct recall, intrusions, new words, and omissions. Participants completed two practice lists at the start of the experiment and took a 15-s break between blocks. Similar to Friedman and Miyake’s study, the dependent variable was the buildup of PI, measured recall accuracy for the first vs. second list of the same category, averaged across blocks. We also calculated the number of list 1 intrusions made during list 2 recall.

Response–Distractor Inhibition Tasks

Flanker Task

In the flanker task (e.g., Eriksen & Eriksen, 1974), participants indicated the identity of the central letter in a string of letters (e.g., KKKHKKK) by pressing one of two buttons on the PsyScope button box. If the central letter was H or K, participants pressed the left button; if the central letter was C or S, participants pressed the right button. In the congruent condition (49.6% of trials), flanking letters were mapped to the same button as the target letter (e.g., HHHHHHH or HHHKHHH). In the neutral condition (21.8% of trials), flanking letters were letters not mapped to a response key (e.g., OOOHOOO). In the incongruent condition (28.6% of trials), flanking letters were mapped to opposite button as the target letter (e.g., SSSHSSS). A single 30-trial practice block preceded four 56-trial experimental blocks with an equal number of trials per condition. The dependent variableswereRT and accuracy for incongruent vs. neutral trials.

Picture–Word Interference Task

In the picture–word interference task (PWI; Lupker, 1979; Schriefers et al., 2002), participants saw a picture with a super-imposed distractor word; participants named the picture while ignoring the word. During the task, each of the 80 pictures was seen in a semantically related condition (i.e., picture and word come from the same category; the interference condition, 50% of trials) and in a semantically unrelated condition (i.e., picture and word come from different categories; the no interference condition, 50% of trials). Written distractor words were not the same as (i.e., did not overlap with) items pictured for naming. The PsyScope button box recorded participant RTs, and the experimenter coded participant responses as correct, incorrect, or voice-key errors. Participants first viewed all of the to-be-named pictures in one practice block where they saw each picture with its correct name. Following this, participants completed 5 practice trials consisting of a previously unseen picture with a super-imposed word, and then two 80-item blocks. The dependent variables were RT and accuracy for semantically related vs. unrelated trials, reflecting the semantic interference effect.

Nonverbal Stroop Task

The nonverbal Stroop task (e.g., Hamilton & Martin, 2005) was administered in 4 blocks of 60 trials. Using buttons on the PsyScope button box, participants indicated the direction an arrow was pointing (right, left), with arrows appearing on either the left side of the screen, the center of the screen, or the right side of the screen. The task contained incongruent trials (left-pointing arrow on the right side of the screen), neutral trials (left-pointing arrow on the center of the screen), and congruent trials (left-pointing arrow on the left side of the screen), with an equal number of trials per condition. A single practice block with 12 trials preceded the experimental blocks. The dependent variableswereRT and accuracy for incongruent vs. neutral trials.

Stroop Task

The Stroop task (Stroop, 1935) was administered in a single block consisting of incongruent, congruent, and neutral stimuli. Each trial was preceded by a beep, followed by target onset. Participants named the color (blue, yellow, orange, red, green, or purple) of the target, which was either a word (in incongruent and congruent conditions) or string of asterisks (in the neutral condition). On incongruent trials (42.2% of trials), color words appeared in a color that was different from the written word (e.g., the word blue written in red ink). On congruent trials (7.8% of trials), color words appeared in the same color as the written word (e.g., the word blue written in blue ink). On neutral trials (50% of trials), the stimulus was a string of colored asterisks. A voice key recorded response times and the experimenter coded participant responses as correct, incorrect, or voice-key errors. One hundred and fifty-four experimental trials were preceded by three practice blocks: the first was for testing and adjusting the microphone’s sensitivity, the second tested color-naming ability by presenting the to-be-named colors as stings of asterisks, and the third consisted of 13 practice trials. The dependent variables were RT and accuracy for incongruent vs. neutral trials, reflecting the Stroop effect.

Processing Speed

Symbol-Digit Coding

The symbol-digit coding task is a paper and pencil task from the MMSE-2 (Folstein et al., 1975, 2001) that was included as a measure of processing speed. The top of the page contained a key, indicating the symbol that corresponded to the numbers 1-9, and participants used this key to code as many numbers as possible within 30 s. The dependent variable was the number of items correctly coded, out of 35. Replicating previous work demonstrating reductions in processing speed with age (e.g., Salthouse, 1994), older adults completed significantly fewer designs than younger adults.

Working Memory

Automated Operation Span Task

The automated operation span task (Ospan) is a measure of WM capacity shown to have good internal consistency and test–retest reliability (Turner & Engle, 1989; Unsworth, Heitz, Schrock, & Engle, 2005). Participants first saw a math operation to verify (true/false), followed by a letter to remember. Following several math operation–letter pairs, participants saw an array of twelve letters with boxes next to them. Participants clicked boxes to indicate the serial order in which letters were presented. For experimental trials, participants completed 3 trials at each set size, with set sizes ranging from 3-7 items. The dependent variable was the operation span defined as the sum of all perfectly recalled sets (Unsworth et al., 2005). This task was presented on a Dell PC running E-Prime (Schneider, Eschman, & Zuccolotto, 2002). Ospan data were missing from two older adults due to experimenter or computer error.

Backwards Digit Span Task

In the backwards digit span from the Weschler Adult Intelligence Scale-Revised (WAIS-R; Weschler, 1981), participants heard a series of numbers presented aurally at a rate of one number/second. Following list presentation, participants recalled numbers in backwards order, starting with the most recently presented item. Participants completed two trials at each list length. All testing started at a list length of 2 items and continued until either errors were made on both trials at a given list length or the maximum list length (8 items) was completed. The dependent variable was the total number of trials correctly recalled.

Sternberg Recognition Task

In the Sternberg recognition task (McElree & Dosher, 1989; Nee & Jonides, 2008), participants saw a list of five serially presented words, with each word presented for 500 ms (similar to Nee & Jonides, 2008). A 300-ms mask was presented over the last word then followed by a probe that remained on the screen for 700 ms. Participants indicated whether the probe word was in the most recently presented list by pressing “yes” or “no” keys. Half of the trials were no trials, and half yes trials. Across the yes trials, each serial position was probed equally often. The dependent variable for this task was overall accuracy across all trials.