Nash Unsworth a, Gene A. Brewer b, Gregory J. Spillers a a Department of Psychology, University of Oregon, Eugene, OR 97403, United States b Department of Psychology, Arizona State University, Tempe, AZ 85287-1104, United States article info Article history: Received 23 September revision received 8 December Available online 20 January 2012 Keywords: Cognitive failure Individual difference Intelligence abstract The present study examined individual differences in everyday cognitive failures assessed by diaries. A large sample of participants completed various cognitive ability measures in the laboratory. Furthermore, a subset of these participants also recorded everyday cognitive failures (attention, retrospective memory, and prospective memory failures) in a diary over the course of a week. Using latent variable techniques the results suggested that individual differences in cognitive abilities (i.e., working memory, attention control, retrospective memory, and prospective memory) were related to individual differences in everyday cognitive failures. Furthermore, everyday cognitive failures predicted SAT scores and partially accounted for the relation between cognitive abilities and SAT scores. These results provide important evidence for individual differences in everyday cognitive failures as well as important evidence for the ecological validity of laboratory cognitive ability measures Elsevier Inc. All rights reserved. Introduction We owe our ability to effectively focus and sustain attention on a task, to retrieve information accurately from memory, and to carryout planned intentions in the future to a well functioning cognitive system. Without question,
this system allows us to carryout the myriad of important and mundane tasks set before us daily. Despite the effectiveness of our overall cognitive system, sometimes we make mistakes resulting in generalized cognitive failures.
For instance, have you ever caught yourself daydreaming during an important meeting Have you ever forgotten the name of a person you were just introduced to Have you ever forgotten to add an attachment to an email before sending it Most people will answer yes to these questions, although the frequency of these cognitive failures likely varies across people. Thus, although we carryout many of our day-to-day tasks successfully, every once in awhile we experience a cognitive failure. Such failures have long been considered an important topic of research in a number of domains including cognitive psychology, cognitive aging, developmental psychology, clinical psychology, educational psychology, neuropsychology, and neuroimaging. An important reason for examining cognitive failures is that not only does the frequency of such errors likely vary as a function of individual differences, neuropsychological disorders, and age, but these failures also have real world consequences. For example, students who are more likely to daydream or mind-wander during lectures may perform more poorly on tests than students who are less likely to mind wander. Furthermore, forgetting to carryout an intention, such as putting the landing gear down before landing, will also have obvious real world consequences (see Reason (fora review. Thus, examining cognitive failures will not only allow fora better understanding of the underlying mechanisms that give rise to such errors but also allow fora better understanding of who is likely to commit such errors. Everyday attention and memory failures Broadly construed, cognitive failures refer to all of the possible different types of failures within the cognitive 0749-596X/$ - see front matter Elsevier Inc. All rights reserved. doi: 10.1016/j.jml.2011.12.005 ⇑ Corresponding author.
email@example.com (N. Unsworth). Journal of Memory and Language 67 (2012) Contents lists available at SciVerse ScienceDirect Journal of Memory and Language journal homepage w w we ls e vie r . com locate j ml div
system (i.e., memorial, attentive, or otherwise) that could conceivably occur. These include lapses of attention, mind wandering, failures of memory, action failures, etc. (e.g., Broadbent, Cooper, FitzGerald, & Parkes, 1982 ). Theoretically, these errors can be conveniently grouped into three classes of failures attention failures, retrospective memory failures, and prospective memory failures (see Heckhausen and Beckman (1990), Norman (1981), and Reason (1984a) for similar taxonomies of action slips. In the current study attention failures refer to situations in which attention could not be maintained and sustained on a task leading to a momentary lapse. Such failures could arise from distracting external stimuli (e.g., aloud noise) or from internal thoughts and distractions (e.g., daydreaming. Thus, these attentional lapses could arise from distractions, from mind wandering, or from absentmindedness (similar to action slips. Retrospective memory failures refer to situations in which information cannot be properly retrieved from the memory system even though that information is likely stored. Retrospective memory failures could include failures over the short-term (e.g., forgetting the name of a person you were just introduced to, failures of autobiographical/personal memory (e.g., forgetting your email password, or failures of more fact-based semantic memory (e.g., forgetting the name of the person who was the President of the United States during the Civil War).
Prospective memory failures refer to situations in which an individual forgets to carryout some intention in the future. For example forgetting to carryout an activity (e.g.,
forgetting to add an attachment to an email, forgetting to do something at a particular time (e.g., forgetting to go to a meeting at 10:15 am, and forgetting to attend an event (e.g., forgetting to go to your sister’s wedding) would all be considered prospective memory failures. Clearly there area number of different ways that the cognitive system can fail and some of these errors can be relatively harmless, whereas other errors could have life-threatening consequences. Understanding these cognitive failure as well as possible sub-classiﬁcations of failures is important in order to not only understand how the cognitive system operates, but it is also for determining who is likely to demonstrate these different failures and in what situations these failures are most likely. That is, an examination of cognitive failures should provide us with more information regarding the underlying cognitive systems that give rise to such errors (attentional and memorial systems) as well as giving some indication of how these systems and their resulting errors are interrelated. An important method for examining everyday attention and memory failures is through diary studies. As the name suggests, in these studies individuals are required to carry a diary for some amount of time and record their attention and memory failures. These studies provide important information about the different types of cognitive failures as well as the relative frequencies with which these cognitive failures occur in everyday life. For example, Reason (1984a) had 63 undergraduates record their action slips in the course of a week. Reason found that many of attentional failures occurred because participants were either preoccupied by internal thoughts or distracted by external stimuli. Furthermore, Reason found that most of these errors occurred during the late afternoon and early evening.
Crovitz and Daniel (had 47 participants record their memory failures. Crovitz and Daniel found that the most frequently occurring memory error was a retrospective memory error (forgetting someone’s name) followed by a prospective memory (forgetting to make a phone call).
Likewise, Terry (examining memory failures in individuals found that prospective memory errors were the most common followed by retrospective memory failures. These results suggest that diary studies provide important information on everyday attention and memory failures. However, little work has examined the relation between cognitive failures assessed with diaries and performance on laboratory tasks. Thus, it is not known whether everyday attention and memory failures reﬂect breakdowns in the same cognitive mechanisms assessed via laboratory tasks and it is not known whether variations in performance on laboratory tasks will be able to predict who is likely to experience everyday attention and memory failures. Individual differences in working memory capacity and cognitive control Theoretically, cognitive failures likely result from general failures in cognitive control. Cognitive control refers to the ability to guide processing and behavior in the service of task goals and this ability is a fundamental aspect of the cognitive system that is thought to be important fora number of higher-level functions. Important components of cognitive control include actively maintaining task goals, selectively and dynamically updating task goals, detecting and monitoring conﬂict, and making adequate control adjustments in the presence of conﬂict (Cohen, As- ton-Jones, & Gilzenrat, 2004 ). These components are thought to inﬂuence processing in a wide range of tasks and situations. As such, the ability to effectively utilize cognitive control and various executive functions (such as updating, switching and inhibition; Miyake et al., should bean important determinant of an individual’s performance in such situations. Early work by Norman (and Reason (a, 1984b)
suggested that cognitive failures arise, in part, due to failures of cognitive control. For example, when attention is disengaged from the current task and focused on other external distracting stimuli or internal thoughts (e.g., daydreaming, cognitive failures are likely to occur. Along this line,
Reason (1984b) sug- gested that susceptibility to cognitive failures appears to be determined by some general control factor that exerts its inﬂuence overall aspects of mental function (p. Theoretically the absence of cognitive control can lead to an increase in the frequency of cognitive failures and this general lack of cognitive control leads to overall increases in all different types of failures rather than speciﬁc failures being due to failures of speciﬁc processing components (i.e., retrospective memory failures as a result of failures in retrospective memory processes. This notion of failures being due to general vs. speciﬁc factors will be examined more thoroughly later. 2 N. Unsworth et al. / Journal of Memory and Language 67 (2012) 1–16
Recently, Engle, Kane and colleagues ( Engle & Kane Kane, Conway, Hambrick, & Engle, 2007 ) have suggested that individual differences in working memory capacity (WMC) are reﬂective of overall differences in general cognitive control abilities that are needed in a host of low-level and high-level cognitive tasks. Evidence for this view comes from a number of studies that have shown that individual differences in WMC are related to individual differences inattention control (e.g., Kane, Bleckley, Conway Engle, 2001; Unsworth & Spillers, 2010 ). Speciﬁcally, prior research has shown that measures of WMC are most pronounced in situations where task goals have to be maintained, particularly in the presence of internal and external distraction as in tasks like the antisaccade, Stroop, ﬂankers, and various vigilance tasks (e.g., Kane et al., 2001;
Unsworth & Spillers, 2010
). Likewise, recent research has suggested that WMC is related to retrospective memory in terms of both encoding and retrieval operations such that individuals high in WMC are better able to encode information and better able to deal with interference at retrieval than individuals low in WMC (e.g., Unsworth, Brewer, & Spillers, 2009; Unsworth & Spillers, 2010 ). Finally, recent work has suggested that WMC is also related to prospective memory in that individuals high in WMC are better able to maintain and carryout planned intentions than low WMC individuals (e.g., Brewer, Knight, Marsh, & Unsworth, 2010 ). Overall, the results from prior studies suggest that high WMC individuals have greater cognitive control abilities than low WMC individuals, and that low WMC individuals are more likely to experience periodic cognitive failures than high WMC individuals in a number of domains. The current research addresses the degree to which these laboratory investigations of cognitive control and attention and memory abilities translates to cognitive failures in individuals day-to-day lives. For the most part, these individual differences studies have examined variation in cognitive control with laboratory tasks and have not examined how individual differences in WMC and cognitive control are related to everyday attention and memory failures. However, one study in particular has examined the relation between individual differences in WMC and cognitive failures in the real world (Kane et al., 2007 ). In this study, Kane et al. had 124 participants perform a battery of WMC tasks in the laboratory and then participants were required to carry personal digit assistants (PDAs) around for 1 week.
During that time the PDA would signal the participant toll out a questionnaire regarding whether they had experienced any mind-wandering (e.g.,
Mason et al., 2007; Small- wood & Schooler, 2006 ) at the time of the signal, thereby providing information on everyday attention failures. Kane et al. found that individual differences in mind wandering were strongly related with measures of WMC, especially during challenging tasks. Speciﬁcally, during challenging tasks low WMC individuals reported more mind-wander- ing than high WMC individuals. This study provides important initial evidence that individual differences in cognitive control as assessed with laboratory tasks are related to everyday attention failures in some contexts. Given the strong theoretical link between individual differences in cognitive control and everyday attention and memory failures, the results from Kane et al. (are exactly what one would expect to see. Those individuals who are better at attention control should demonstrate less mind wandering (and fewer lapses) than those individuals who are poorer at attention control. Indeed, in discussing potential reasons for individual differences in cognitive failures Reason and Mycielska (suggested that the liability to minor cognitive failures may not necessarily depend primarily upon the quantity of the sparse attentional resource, but upon the facility with which individuals dispose of this commodity from moment to moment. In other words, people may vary in the effectiveness with which they deploy their attentional reserve in response to the current demands of the situation (p. Thus, individual differences inattention control should predict who is likely to experience frequent cognitive failures. Although the results of
Kane et al. (suggest this might be the case, it should be noted that Kane et al. only examined individual differences in WMC and mind-wandering and did not examine the other cognitive failures or other cognitive abilities (such as attention control, retrospective memory, or prospective memory).
Clearly more work is needed to examine the potential linkages between cognitive control in the laboratory and cognitive failures in everyday life. A further reason to explore individual differences in everyday cognitive failures is because laboratory assessments of WMC have long been shown to predict performance on scholastic ability measures such as the SAT ( Engle, Tuholski, Laughlin, & Conway, 1999 ) which can be seen as a proxy for general intelligence (Frey & Detterman, 2004 ). The fact that individual differences in cognitive control (as assessed by WMC) are related to individual differences in scholastic abilities and intelligence has long been taken as evidence for the general importance of cognitive control and much research has been devoted to examining the reasons for this relation. However, as yet, no research has directly examined the extent to which everyday cognitive failures are important predictors of scholastic abilities and general intelligence and the extent to which everyday cognitive failures account for the relation between cognitive abilities assessed in the laboratory and formal intellectual tasks such as the SAT. Theoretically, those individuals with poor cognitive control capabilities should be more likely to experience lapses of attention (e.g., mind-wander- ing during test taking) and memory (e.g., momentarily forgetting the deﬁnition of the word diatribe) which could impact their scores on formal intellectual tasks. If this is the case, then individual differences in everyday cognitive failures should be related to performance on scholastic and intellectual tasks such as the SATs. The present study The goal of the present study was to better examine individual differences in cognitive failures. In particular,
three main questions were examined. First, do people who differ in cognitive abilities differ in the extent to which they experience real-world cognitive failures Spe- ciﬁcally, do measures of attention control, WMC, retrospective memory,
and prospective memory predict N. Unsworth et al. / Journal of Memory and Language 67 (2012) 1–16 3
everyday cognitive failures Intuitively and theoretically one would expect that attention and memory processes assessed in the laboratory would predict attention and memory failures in everyday life, but, little work has directly examined these issues. Thus, it is unclear whether individual differences in cognitive abilities and cognitive control more speciﬁcally will predict cognitive failures in everyday life. Indeed, as noted by Reason and Mycielska (laboratory tests do not always prove to be good predictors of a person’s performance in the outside world. The attempt to bridge this gulf between the highly speciﬁc and easily manipulable laboratory task and the uncontrollable and myriad concerns of actual day-to-day living remains one of psychology’s most difﬁcult challenges. Both modes of inquiry are necessary and important but reconciling their ﬁndings is not easy (p. 233). We agree with Reason and Mycielska that attempting to bridge laboratory studies of cognition with ecologically valid studies of everyday attention and memory failures is an important endeavor and one that will help us to not only understand the potential mechanisms that give rise to various cognitive failures, but also allow us to predict who is likely to experience frequent cognitive failures in everyday life. Furthermore, most prior studies of everyday cognitive failures have speciﬁcally examined attention failures (e.g., lapses of attention and slips of action) or memory failures
(retrospective and prospective failures, but little work has been done to examine the extent to which different types of cognitive failures are related to one another. Thus, it unclear whether individual differences in cognitive failures represent a global issue or whether various cognitive failures (attention failures, retrospective memory failures,
and prospective memory failures) represent distinct constructs. Intuitively it seems plausible that individual differences in cognitive failures represent abroad construct such that a person with poor cognitive control is likely to experience many cognitive failures across the various different subcategories of errors. Indeed, Reason and Mycielska (1982) suggested that individual differences in cognitive failures were due to a general problem and were not localized to speciﬁc domains (i.e., attention, memory, etc.). Thus, based on this theory one would expect that a single factor represents all cognitive failures. It also seems plausible that individual differences in cognitive failures are relatively distinct such that a person who experiences many lapses of attention does not necessarily experience many retrospective memory failures, thus suggesting that various cognitive failures are distinct. That is, to the extent that attention control, retrospective memory, and prospective memory are relatively distinct (but related) constructs, one would expect some relations between the different types of cognitive failures that arise from each, but that these failures would be directly tied to the underlying construct rather than to a general factor. Therefore, the second main question addressed in the current study was to what extent are different types of cognitive failures (e.g., attention failures, retrospective memory failures, and prospective memory failures) related and to what extent do they represent general or speciﬁc constructs?
Finally, the third main question addressed in the current study was to what extent are everyday cognitive failures related to scholastic abilities and intelligence as indicated by SAT scores Speciﬁcally, prior research has found that individual differences in cognitive abilities assessed in the laboratory (such as WMC) strongly predict
SAT scores (e.g., Engle et al., 1999 ), but prior research has not addressed the extent to which everyday attention and memory failures predict variation in SAT scores. Theoretically, individuals with lower cognitive control abilities who also experience more cognitive failures should also have lower scholastic abilities as a lapse of attention or being distracted while taking a test could potentially lead to a lower than normal score. If this is the case then individual differences in everyday cognitive failures should be related to scholastic abilities as measured by the SATs. To address these questions we a tested a large number of participants (Non several laboratory tasks thought to measure WMC, attention control, retrospective memory, and prospective memory. A subset of participants (N 100) also agreed to carry diaries fora week in which they recorded cognitive failures (attention, retrospective memory, and prospective memory failures) they experienced each day. Finally, we obtained verbal and quantitative SAT scores for each participant via self- report.
The current study goes signiﬁcantly beyond prior work by examining a large number of participants on a number of different laboratory measures of cognitive abilities as well as diary self-reports of a number of different types of cognitive failures. We used a latent variable approach to examine the relations between the laboratory cognitive measures and the various everyday cognitive failures measures. In order to derive latent variables for the constructs of interest, multiple indicators of each cognitive construct were used. This was done in order to ensure that any lack of a relation found between laboratory cognitive measures and everyday cognitive failures would not be due to unreliability or idiosyncratic task effects. Therefore, multiple measures of each cognitive construct were used to create latent variables of WMC, attention control, retrospective memory, and prospective memory and examine the relation between these latent variables with the self-report measures. By examining a large number of participants and a large and diverse number of measures we should be able to better characterize individual differences in everyday cognitive failures and address our three questions of primary interest.
Method Participants A total of 165 participants (68% female) were recruited from the subject-pool at the University of Georgia. Participants were between the ages of 18 and 35 (M SD 1.51) and received course credit for their participation. Each participant was tested individually in a laboratory session lasting approximately 2 h. Of these participants, 100 agreed to carry diaries fora week in which they recorded cognitive failures. 4 N. Unsworth et al. / Journal of Memory and Language 67 (2012) 1–16
Materials and procedure After signing informed consent, all participants completed operation span, symmetry span, reading span, free recall, antisaccade, low association cue–target PM, paired associates recall, arrow ﬂankers, non-focal PM. All tasks were administered in the order listed above. Following the task, participants who were willing to participate in the diary portion of the study were given explicit and elaborate instruction on the diaries. Laboratory tasks Working memory capacity (WMC) tasks Operation span (Ospan). Participants solved a series of math operations while trying to remember a set of unrelated letters (F, H, J, KL, NP, QR, STY. Participants were required to solve a math operation and after solving the operation they were presented with a letter for 1 s. Immediately after the letter was presented the next operation was presented. Three trials of each list-length (3–7) were presented fora total possible of 75. The order of list-length varied randomly. At recall, letters from the current set were recalled in the correct order by clicking on the appropriate letters (see Unsworth, Heitz, Schrock, and Engle
for more details. Participants received three sets (of list-length two) of practice. For all of the span measures, items were scored if the item was correct and in the correct position. The score was the proportion of correct items in the correct position. Symmetry span (Symspan). In this task participants were required to recall sequences of red squares within a matrix while performing a symmetry-judgment task. In the sym- metry-judgment task participants were shown an 8 matrix with some squares ﬁlled in black. Participants decided whether the design was symmetrical about its vertical axis. The pattern was symmetrical half of the time. Immediately after determining whether the pattern was symmetrical, participants were presented with a 4 4 matrix with one of the cells ﬁlled in red forms. At recall, participants recalled the sequence of red-square locations in the preceding displays, in the order they appeared by clicking on the cells of an empty matrix (see Unsworth, Re- dick, Heitz, Broadway, and Engle (for more details). There were three trials of each list-length with list-length ranging from 2 to 5 fora total possible of 42. The same scoring procedure as Ospan was used. Reading span (Rspan).
Participants were required to read sentences while trying to remember the same set of unrelated letters as Ospan. For this task, participants read a sentence and determined whether the sentence made sense or not (e.g. The prosecutor’s dish was lost because it was not based on fact. ?’’). Half of the sentences made sense while the other half did not. Nonsense sentences were made by simply changing one word (e.g. dish from case) from an otherwise normal sentence. Participants were required to read the sentence and to indicate whether it made sense or not. After participants gave their response they were presented with a letter for 1 s. At recall, letters from the current set were recalled in the correct order by clicking on the appropriate letters (see
Unsworth et al. (for more details. There were three trials of each list-length with list-length ranging from 3 to 7 fora total possible of. The same scoring procedure as Ospan was used. Attention control (AC) tasks Antisaccade. In this task (Kane et al., 2001 ) participants were instructed to stare at a ﬁxation point which was onscreen fora variable amount of time (200–2200 ms. A ashing white ‘‘=’’ was then ﬂashed either to the left or right of ﬁxation (of visual angle) forms. This was followed by a 50 ms blank screen and a second appearance of the cue forms making it appear as though the cue) ﬂashed onscreen. Following another 50 ms blankscreen the target stimulus (a BP, or R) appeared onscreen forms followed by masking stimuli (an H forms and an 8 which remained onscreen until a response was given). All stimuli were presented in Courier New with a 12 point font. The participants task was to identify the target letter by pressing a key for BP, or R (keys 1, 2, or 3 on the number keypad) as quickly and accurately as possible. In the prosaccade condition the ﬂashing cue (=) and the target appeared in the same location. In the antisaccade condition the target appeared in the opposite location as the ﬂashing cue. Participants received, in order, 10 practice trials to learn the response mapping, 10 trials of the prosaccade condition, and 40 trials of the antisaccade condition. Arrow ﬂankers. Participants were presented with a ﬁxation point forms. This was followed by an arrow directly above the ﬁxation point forms. The participants task was to indicate the direction the arrow was pointing
(pressing the F for left pointing arrows and pressing J for right pointing arrows) as quickly and accurately as possible. On 30 neutral trials the arrow was ﬂanked by two horizontal lines on each side. On 30 congruent trials the arrow was ﬂanked by two arrows pointing in the same direction as the target arrow on each side. Finally, on 30 incongruent trials the target arrow was ﬂanked by two arrows pointing in the opposite direction as the target arrow on each side.
All trial types were randomly intermixed. The dependent variable was the reaction time difference between incon- gruent and congruent trials. Psychomotor vigilance task (PVT). The psychomotor vigilance task ( Dinges & Powell, 1985 ) was used as the primary measure of sustained attention. Participants were presented with a row of zeros onscreen and after a variable amount of time the zeros began to count up in 1 ms intervals from 0 ms. The participants task was to press the spacebar as quickly as possible once the numbers started counting up. After pressing the spacebar the RT was left onscreen for 1 s to provide feedback to the participants. Interstimulus intervals were randomly distributed and ranged from 1 to 10 s. The entire task lasted for 10 min for each individual (roughly 75 total trials. The dependent variable was the average reaction time for the slowest of trials ( Dinges & Powell, N. Unsworth et al. / Journal of Memory and Language 67 (2012) 1–16 5