Event-Based Prospective Memory Among Veterans: the Role of Posttraumatic Stress Disorder Symptom Severity in Executing Intentions

Journal of Clinical and Experimental Neuropsychology

ISSN: 1380-3395 (Print) 1744-411X (Online) Journal homepage: http://www.tandfonline.com/loi/ncen20

Event-based prospective memory among veterans: The role of posttraumatic stress disorder symptom severity in executing intentions

Craig P. McFarland, Justin B. Clark, Lewina O. Lee, Laura J. Grande, Brian P. Marx & Jennifer J. Vasterling

To cite this article: Craig P. McFarland, Justin B. Clark, Lewina O. Lee, Laura J. Grande, Brian P. Marx & Jennifer J. Vasterling (2016) Event-based prospective memory among veterans: The role of posttraumatic stress disorder symptom severity in executing intentions, Journal of Clinical and Experimental Neuropsychology, 38:2, 251-260, DOI: 10.1080/13803395.2015.1102203

To link to this article: http://dx.doi.org/10.1080/13803395.2015.1102203

Published online: 21 Dec 2015.

Submit your article to this journal

Article views: 11

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ncen20

Download by: [University of Montana] Date: 08 January 2016, At: 10:20 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY, 2016 VOL. 38, NO. 2, 251–260 http://dx.doi.org/10.1080/13803395.2015.1102203

Event-based prospective memory among veterans: The role of posttraumatic stress disorder symptom severity in executing intentions Craig P. McFarlanda,b, Justin B. Clarkc, Lewina O. Leec,d,e, Laura J. Grandea,d,f, Brian P. Marxa,c,d and Jennifer J. Vasterlinga,c,d aPsychology Service, VA Boston Healthcare System, Boston, MA, USA; bDepartment of Psychology, University of Montana, Missoula, MT, USA; cDepartment of Veterans Affairs National Center for Posttraumatic Stress Disorder, VA Boston Healthcare System, Boston, MA, USA; dBoston University School of Medicine, Boston, MA, USA; eBoston University School of Public Health, Boston, MA, USA; fTranslational Research Center for TBI and Stress Disorders (TRACTS), VA Boston Healthcare System, Boston, MA, USA

ABSTRACT ARTICLE HISTORY Objective: Posttraumatic stress disorder (PTSD) has been linked with neuropsychological Received 2 July 2015 deficits in several areas, including attention, learning and memory, and cognitive Accepted 25 September 2015 inhibition. Although memory dysfunction is among the most commonly documented KEYWORDS deficits associated with PTSD, our existing knowledge pertains only to retrospective Prospective memory; memory. The current study investigated the relationship between PTSD symptom posttraumatic stress severity and event-based prospective memory (PM). Method: Forty veterans completed disorder; event-based; a computerized event-based PM task, a self-report measure of PTSD, and measures of anxiety retrospective memory. Results: Hierarchical regression analysis results revealed that PTSD symptom severity accounted for 16% of the variance in PM performance, F(3, 36) = 3.47, p < .05, after controlling for age and retrospective memory. Additionally, each of the three PTSD symptom clusters was related, to varying degrees, with PM performance. Conclusions: Results suggest that elevated PTSD symptoms may be associated with more difficulties completing tasks requiring PM. Further examination of PM in PTSD is warranted, especially in regard to its impact on everyday functioning.

Associations between posttraumatic stress disorder hippocampus, and amygdala (Rauch, Shin, & (PTSD) and deficits on performance-based neu- Phelps, 2006). In addition to their mechanistic ropsychological tasks have been well documented significance, neurocognitive deficits may contri- and commonly include difficulty learning and bute to the functional impairments that commonly recalling new information (Brewin, 2011; accompany PTSD (Hoge, Terhakopian, Castro, Schuitevoerder et al., 2013), attentional deficits Messer, & Engel, 2007; Rodriguez, Holowka, & (Jenkins, Langlais, Delis, & Cohen, 2000), impaired Marx, 2012) and adversely affect quality of life Downloaded by [University of Montana] at 10:20 08 January 2016 cognitive inhibition (Shucard, McCabe, & (Schnurr, Lunney, Bovin, & Marx, 2009). Szymanski, 2008; Swick, Honzel, Larsen, Ashley, Memory dysfunction holds particular signifi- & Justus, 2012), and working memory deficits cance to PTSD, as it ranks high among the cogni- (Leskin & White, 2007; Vasterling, Brailey, tive domains commonly impaired in PTSD and Constans, & Sutker, 1998). The results of a recent has relevance to neuroanatomical models of meta-analysis support these findings, further indi- PTSD implicating the hippocampus and prefrontal cating that the largest effect sizes pertained to new cortex as well as to day-to-day functioning verbal learning and processing speed (Scott et al., (Geuze, Vermetten, de Kloet, Hijman, & 2015). Such deficits are thought to reflect under- Westenberg, 2009). Most studies of memory per- lying neural abnormalities associated with PTSD, formance in PTSD (e.g., Gilbertson, Gurvits, including dysfunction within limbic and paralim- Lasko, Orr, & Pitman, 2001; Jelinek et al., 2006; bic circuits involving the prefrontal cortex, Vasterling et al., 2002) have used routine clinical

neuropsychological episodic memory tasks. These performance. More specifically, we predicted that tasks target “retrospective memory,” or the ability the Diagnostic and Statistical Manual of Mental to explicitly recall information or events that were Disorders, Fourth Edition (DSM–IV; American previously learned or experienced. Less is known Psychiatric Association, 1994) PTSD symptom about the association between PTSD and prospec- severity would be inversely correlated with perfor- tive memory (PM), or the ability to remember to mance on a PM task after adjusting for both age complete a plan of action in the future. Examples and retrospective memory performance. Because of activities that require PM include remembering prior research has indicated that PTSD symptom to pick a child up from school, attend a meeting, or clusters may be differentially correlated with neu- take medication. Several studies have revealed a ropsychological performance (Swick et al., 2012; relationship between PM measures and day-to- Vasterling et al., 1998), we also explored the extent day functioning, including medication adherence to which specific DSM–IV PTSD symptom clusters (Vedhara et al., 2004) and activities of daily living (i.e., reexperiencing, avoidance/numbing, arousal) (Smits, Deeg, & Jonker, 1997). The purpose of this were associated with PM performance. Finally, study was to extend our understanding of memory because a growing body of research has linked deficits associated with PTSD by examining the depressive symptoms, which are frequently comor- extent to which PTSD symptom severity is related bid with PTSD (Campbell et al., 2007), to impaired to PM, a functionally relevant domain not pre- PM performance (Li, Loft, Weinborn, & Maybery, viously examined in the PTSD literature. 2014; Rude, Hertel, Jarrold, Covich, & Hedlund, There are a number of reasons to hypothesize 1999), we investigated the relationship between that PTSD would be associated with impaired PM. depressive symptoms and PM. First, neuroanatomical models of PTSD (e.g., Rauch et al., 2006) implicate the hippocampus and medial prefrontal cortex, areas thought to be Method involved in PM (Adda, Castro, Além-Mar e Silva, Participants de Manreza, & Kashiar, 2008; Burgess, Gonen- Yaacovi, & Volle, 2011; Burgess, Quayle, & Frith, Forty military veterans were recruited from PTSD- 2001). Relatedly, PTSD is associated with deficits related participant recruitment databases (i.e., in retrospective memory (Johnsen & Asbjørnsen, veterans who participated in previous studies giving 2008) and inhibitory processing (Shucard et al., their permission to be recontacted for future 2008; Swick et al., 2012), both of which are thought research) at the Veterans Affairs Boston to be critical for successful PM (Glisky, 1996; Healthcare system. These databases were developed Martin, Kliegel, & McDaniel, 2003; Schnitzspahn, for a range of military trauma studies in which Stahl, Zeintl, Kaller, & Kliegel, 2013) and stem PTSD was a variable of interest, including studies from decreased activity in the hippocampus and incorporating healthy non-PTSD comparison inferior lateral prefrontal cortex (Rauch et al., groups. Veterans were excluded from participating Downloaded by [University of Montana] at 10:20 08 January 2016 2006), respectively. Finally, studies have also if they (a) endorsed a history of head trauma with an revealed an association between PM deficits and associated loss of consciousness of greater than 15 anxiety states (Harris & Cumming, 2003; Harris & min, (b) endorsed central nervous system disease, Menzies, 1999), behaviors (Cuttler & Graf, 2007, or (c) were taking neuroleptic medications. 2008, 2009), and disorders, specifically obsessive Additionally, volunteers were excluded from parti- compulsive disorder (Marsh et al., 2009; cipating if they met criteria for current (defined as Racsmány, Demeter, Csigó, Harsányi, & Németh, past three months) alcohol or substance abuse or 2011; although see Harris, Vaccaro, Jones, & Boots, dependence or lifetime psychotic or bipolar disor- 2010, for contradictory findings). Thus, it is con- der as assessed via an abbreviated Structured ceivable that PM is impaired in other disorders Clinical Interview for DSM–IV Axis I Disorders with prominent anxiety features, such as PTSD. (SCID; First, Spitzer, Gibbon, & Williams, 2002). In this first study examining the relationship Finally, participants were excluded if they endorsed between PM and PTSD, we hypothesized that a history of attention deficit hyperactivity disorder PTSD symptom severity, as indicated by scores or learning disability in response to screening ques- on a self-report measure of PTSD symptomatol- tions. One hundred and fifty veterans were initially ogy, would be inversely related to PM contacted and screened by phone. Of those 150 JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 253

veterans, 89 were excluded for screening positive for characterization of the sample we also used a a current alcohol use disorder on the Alcohol Use BDI–II cut-score of 20 (Li et al., 2014) to identify Disorders Identification Test, seizure disorder, those participants that endorsed either moderate attention-deficit/hyperactivity disorder (ADHD), or severe depressive symptoms. As an estimate of learning disability, history of traumatic brain injury, native intellectual potential we administered the or history of psychiatric disorder. All veterans that Wechsler Test of Adult Reading (WTAR; met study criteria and agreed to participate pro- Wechsler, 2001). Retrospective declarative memory vided written informed consent and were compen- was assessed using the 16-item California Verbal sated for participating in the experiment. All data Learning Test–II (CVLT–II; Delis, Kramer, reported were collected in compliance with the Kaplan, & Ober, 2000) Long Delay Free Recall Helsinki Declaration and with the approval of the trial. As an additional index of retrospective mem- Department of Veterans Affairs Internal Review ory, following the completion of the PM task, Board. participants were asked to describe the tasks involved in the study. PTSD symptoms Procedure PTSD symptom severity was assessed using the PTSD Checklist, Civilian Version (PCL-C; The experimental session lasted approximately 90 Weathers, Litz, Herman, Huska, & Keane, 1993). min. After providing written informed consent and The PCL-C is a 17-item self-report scale that completing an abbreviated SCID, participants assesses DSM–IV symptom criteria over a period completed the computer-based PM task. of 1 month. The response set for each item ranges Participants then completed the WTAR and the from 1 (“not at all”)to5(“extremely”), yielding a CVLT–II, followed by the self-report question- potential summary score range of 17–85, with naires. Finally, participants were debriefed and higher scores indicating greater PTSD symptom provided time to ask questions. severity. The PCL has demonstrated acceptable levels of discriminant validity compared with PM paradigm other measures of psychopathology (Blanchard, Jones-Alexander, Buckley, & Forneris, 1996) and PM paradigms are designed to simulate routine is highly correlated with the Clinician- everyday tasks in which individuals are typically Administered PTSD Scale (Blake et al., 1995). As involved in one or more ongoing tasks throughout described below, in our primary analyses, we the course of a day that must be interrupted at the examined PTSD as a dimensional, rather than appropriate time to initiate a prospectively remem- categorical, variable as a reflection of our interest bered activity (e.g., remembering to take a medica- in the continuous nature of the construct and tion on schedule while performing routine work relatedly to test our hypotheses that PM would tasks). In everyday life, the task to be remembered Downloaded by [University of Montana] at 10:20 08 January 2016 show a linear relation to PTSD within the entire may optimally occur in response to a specific event continuum of symptom severity. To provide (e.g., passing the grocery store on the way home) further clinical characterization of the sample, in or within a specific timeframe (e.g., taking a med- secondary analyses we used a well-established PCL ication in the evening). The PM paradigm in this cut-score (50) to indicate clinically significant study was a classic laboratory event-based task (vs. PTSD symptomatology (Karstoft, Andersen, time-based) in that participants were engaged in Bertelsen, & Madsen, 2014). an ongoing distractor task requiring periodic inter- ruption in response to cues (i.e., “events”) to successfully complete a specified PM, or “target,” task Depression symptoms, intellectual potential, (for similar paradigms see Altgassen, Henry, and retrospective memory Bürgler, & Kliegel, 2011; Loft et al., 2014). This We administered the Beck Depression Inventory– task has been shown to be a reliable measure of II (BDI–II; Beck, Steer, & Brown, 1996) to assess PM (Kelemen, Weinberg, Alford, Mulvey, & depression symptom severity. As with PTSD, we Kaeochinda, 2006). also examined depression as a dimensional vari- Specifically, participants were given a sheet of able. However, to provide additional clinical paper detailing task instructions and were asked to 254 C. P. McFARLAND ET AL.

read it carefully. The instructions indicated that presses made after the presentation of subsequent they would be asked to complete a multiple-choice questions. test of general knowledge and trivia (i.e., the Additional variables of interest included mea- ongoing task) and that they would have a total of sures of ongoing task performance: the number of 12 seconds to respond to each question. trivia questions answered correctly and the average Participants were also informed that, during the response time to answer each question. trivia task, they should press the “6” key in Performance differences on these measures could response to any trivia questions that contained reflect differences in engagement with the ongoing the word “president” (i.e., the secondary PM task and subsequently contribute to PM perfor- task). The instructions for the PM task were then mance differences. Therefore, we measured presented on the computer monitor. Participants ongoing task performance to determine whether were then asked to write that exact instruction on a any PM performance deficits were explainable on sheet of paper and to repeat the instructions aloud the basis of task engagement. to ensure that they each understood what they were to do in the experiment. Participants were Data analysis next told that the experiment would begin, and no further mention of the PM task was made. All independent and dependent variables were Participants did not receive feedback regarding normally distributed, with skewness ranging from PM task performance during the experiment. –0.75 to 0.55, and kurtosis ranging from –1.4 to – The multiple-choice trivia questions with the 0.1. To test our primary hypothesis, that PTSD embedded PM task proceeded for approximately symptom severity would be inversely related to 42 minutes, after which participants were asked to PM performance, we conducted a hierarchical describe the tasks involved (i.e., answering multi- regression analysis. Age was the first variable ple-choice questions; pressing the “6” key when entered, followed by CVLT–II Long Delay Free seeing the word “president”). The trivia questions Recall. This order was based on previous work consisted of 196 questions selected from a previous showing that both age and retrospective memory study (McDaniel, Glisky, Rubin, Guynn, & can affect PM performance (McDaniel & Einstein, Routhieaux, 1999) and were presented on a 2011). The PCL-C summary score was entered Gateway E-6610D PC, using DMDX (Forster & next to identify any unique variance contributed Forster, 2003). Questions were presented in ran- by PTSD symptom severity to PM performance dom order in Times New Roman 12-point font in over and above that explained by age and retro- the center of the monitor. Beneath each question, spective memory. For ease of interpretation, age four response options corresponding to multiple- and CVLT–II Long Delay Free Recall were choice answers were presented and labeled “A,” centered at the sample mean; the PCL-C was cen- “B,”“C,” or “D.” Successive questions appeared at tered at 50 and divided by 10 such that the result- 12-second intervals, regardless of how quickly the ing regression coefficients would reflect the Downloaded by [University of Montana] at 10:20 08 January 2016 participant responded to the question. The dura- associated change in PM performance for every tion of the interval between making a response and 10-point difference in PCL-C scores, an increment the presentation of the next question therefore reflecting a clinically significant increase in PTSD varied depending on the response time. symptomatology. Performance on the eight target PM trials, indi- We considered including BDI–II scores in the cated by the total number of correct responses, model to examine the association of PTSD with comprised the primary dependent variable. PM, independent of depressive symptoms, given Correct PM responses were defined as pressing the high comorbidity of depression with PTSD; the “6” key in response to questions that contained however, examination of multicollinearity indi- the word “president,” either during the presenta- cated that the variance inflation factor (VIF) for tion of the target question or immediately follow- BDI scores was unacceptably high (>2.5) when ing its presentation, but prior to the presentation included in the same model as the PCL-C, which of a subsequent question. All “6” key presses made could lead to unstable regression coefficients. in response to “president” and prior to the appear- Thus, we examined associations between depres- ance of subsequent questions were given a value of sive symptoms and PM performance in a second, 1. A value of 0 was given for any omissions or key post hoc model that was identical to the first JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 255

regression model with respect to the variables of With regard to retrospective memory, the overall interest and the order of entry, except that BDI–II sample performed in the average range and summary scores were substituted for PCL-C sum- recalled a mean of 12.2 of 16 potential items on mary scores in the third step. The post hoc model the CVLT–II Long Delay Free Recall (SD = 2.8). allowed us to compare the magnitude of the asso- Table 1 provides additional information related to ciation between depression and PM and that self-report measure data and task performance. between PTSD and PM. Partial correlational analyses controlling for age Prospective memory performance and retrospective memory were conducted to explore potential relations between PCL-C symp- Tests of multicollinearity for each regression ana- tom cluster scores and PM performance. Finally, lysis indicated that (a) all VIFs ranged from 1.0 to Pearson product–moment correlations were con- 1.3, suggesting that the predictor variables in each ducted to identify significant associations between model were not collinear with one another; and (b) indicators of task engagement (i.e., number of trivia Durbin–Watson statistics for both models were questions answered correctly and reaction times to also acceptable (2.0 to 2.2), indicating that the trivia questions) and PTSD symptom severity. assumption of independent errors was met. As depicted in Table 2, the primary regression analysis examining the association between PTSD Results symptom severity and PM revealed that age accounted for 5% of the variance in PM perfor- Participant characteristics mance, with CVLT–II Long Delay Free Recall con- The majority of participants were White (70%) and tributing an additional 1% (ps > .10). PTSD male (88%), with a mean age of 39.6 years (SD = symptom severity, as indicated by PCL-C scores, 10.2 years; range = 24–60 years). Veterans had accounted for an additional 16% of the variance in completed an average of 13.9 years of education PM performance. As hypothesized, more severe (SD = 2.1 years) and obtained a mean WTAR score PTSD symptoms were associated with poorer PM of 38.3 (SD = 7.2), suggesting that as a group, performance. The beta coefficient for PCL-C (B = – participants were of average intelligence. Mean 0.71) indicates that every 10 additional points on PCL-C and BDI–II scores were 45.9 (SD = 17.0), the PCL-C was associated with a PM performance and 16.0 (SD = 12.3), respectively, indicating a decrement of 0.71. Overall, the model accounted for broad range of both PTSD and depression severity 22% of the variance in PM performance, F(3, 36) = levels and that the sample on average reported 3.47, p < .05. These findings held when the above moderate PTSD and mild depression symptoms. regression was repeated with PTSD symptoms

Table 1. Demographic information, self-report, and task performance. Downloaded by [University of Montana] at 10:20 08 January 2016 Variable M (SD) Range n (%) Age (years) 39.6 (10.2) 24–60 Education (years) 13.9 (2.1) 11–19 Women 5 (12.5) Ethnic minority 5 (12.5) Estimated IQ (WTAR) 107.2 (11.0) 79–125 PCL-C summary score 45.9 (17.0) 17–84 PTSD screen positive 19 (47.5) BDI–II summary score 16.0 (12.3) 0–46 Depression screen positive 14 (35) CVLT–II Trial 1 7.2 (2.8) 2–14 Trials 1–5 55.4 (12.1) 29–75 Long Delay Free Recall 12.2 (2.8) 6–16 PM accuracy (out of 8) 4.2 (2.8) 0–8 Ongoing task performance Hit rate .37 (.07) .24–.54 Reaction time (ms) 2417.42 (907.14) 1059.52–5270.45 Note. PTSD = posttraumatic stress disorder; PCL-C = PTSD Checklist, Civilian Version; BDI–II = Beck Depression Inventory–II; WTAR = Wechsler Test of Adult Reading; CVLT–II = California Verbal Learning Test–Second Edition; PM = prospective memory. A positive PTSD screen = PCL-C ≥ 50. A positive depression screen = BDI–II ≥ 20. 256 C. P. McFARLAND ET AL.

Table 2. Multiple regression analysis with age, retrospec- Table 3. Partial correlation between PCL-C clusters, BDI–II, tive memory, and PCL-C as predictors of PM performance. and PM performance controlling for age and retrospective Variable BSEB β memory. Step 1 Variable PCL B PCL C PCL D BDI–II Constant 4.17 0.45 PM –.31+ –.43** –.41* –.33* Age –0.06 0.04 –.21 PCL B — .73*** .75*** .60*** Step 2 PCL C — .90*** .75*** Constant 4.17 0.45 PCL D — .75*** Age –0.04 0.05 –.16 BDI–II — – CVLT II LDF 0.13 0.18 .13 Note Step 3 . PM = prospective memory; PTSD = posttraumatic stress dis- Constant 3.80 0.44 order; PCL-C = PTSD Checklist, Civilian; PCL B = PCL-C Cluster B Age –0.02 0.05 –.07 (reexperiencing); PCL C = PCL-C Cluster C (avoidance/numbing); – PCL D = PCL-C Cluster D (hyperarousal); BDI–II, Beck Depression CVLT II LDF 0.07 0.17 .06 – PCL-C –0.71 0.25 –.43** Inventory II . +p = .06. *p ≤ .05. **p ≤ .01. ***p ≤ .005 or p ≤ .001. Note. Age and retrospective memory were centered at the sample mean, and PCL-C was centered at 50 and divided by 10. PM = prospective memory; CVLT–II LDF = California Verbal Learning Test–Second Edition, Long Delay Free Recall; PTSD = posttraumatic with less proficient PM performance (r = –.43, stress disorder; PCL-C = PTSD Checklist, Civilian. R2 = .05 for Step 1; ΔR2 = .01 for Step 2; ΔR2 = .17 for Step 3. p < .01; r = –.41, p = .01, respectively). Though **p ≤ .01. not statistically significant, the partial correlation between Cluster B (reexperiencing symptoms) scores and PM performance (r = –.31, p = .06) entered as a dichotomous variable with a cut- suggests a trend for more severe reexperiencing score of 50, F(3, 36) = 2.72; p = .05. Additional symptoms to be associated with less proficient analyses were conducted to determine whether PM performance. The results of a Fisher’s r-to-z native intellectual potential (WTAR), education, transformation confirmed that the correlations basic attention (CVLT–II Trial 1), or learning between PM performance and each of the three (CVLT–II total) accounted for greater variance symptom clusters were not significantly different than retrospective memory and altered the rela- (ps > .28). tion between PTSD and PM. We substituted those variables, independently, for CVLT–II Long Delay Free Recall in the primary regres- Ongoing task performance sion. The results were unchanged and revealed that native intellectual potential, education, Correlational analyses revealed no significant asso- attention, and learning each accounted for non- ciation between PCL-C scores and the number of significant portions of variance in PM perfor- trivia questions answered correctly (r = .13; p > mance and had no impact on the relation 0.5). Similarly, PCL-C summary scores did not between PTSD and PM. correlate significantly with reaction times when – The results of a second, post hoc regression, answering trivia questions (r = .04; p > .9). in which BDI–II scores were substituted for These results suggest that participants did not dif- Downloaded by [University of Montana] at 10:20 08 January 2016 PCL-C scores, revealed that the inclusion of fer in their level of engagement as a function of depressive symptoms accounted for an addi- symptom severity. tional 10% of the variance in PM performance (β = –.35, p < .05) over and above that Retrospective memory accounted for by age and CVLT–II Long Delay Free Recall; the overall model approached but All participants successfully recalled the speci- did not reach statistical significance, F(3, 36) = fics of the task, including that they were to 2.34, p = .09. answer trivia questions and to press the “6” Partial correlations, depicted in Table 3, key whenever the word “president” appeared. revealed that the three PTSD symptom clusters These results, in addition to the failure were each related, to varying degrees, to PM (reported previously) to detect a significant performance. Specifically, Clusters C (avoidance relation between CVLT–II Long Delay Free and numbing symptoms) and D (hyperarousal Recall and PM performance, suggest that retro- symptoms) were each negatively correlated with spective memory impairment cannot explain the PM performance, such that higher levels of both association between PTSD symptoms and PM avoidance and arousal symptoms were associated performance. JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 257

Discussion be learned and maintained) and by the finding that, as compared with normative data, no partici- This study examined the relation between PTSD pants performed in the impaired range on a mea- symptom severity and PM in a sample of military sure of delayed memory (CVLT–II Delayed veterans. Although PTSD has been associated with Free Recall). The finding that each participant a variety of neuropsychological abnormalities, successfully recalled the cue–intention association including attentional deficits (Jenkins et al., after completing the study further suggests that the 2000), diminished cognitive inhibition (Shucard poor PM performance in this study was not driven et al., 2008; Swick et al., 2012), and working mem- by retrospective memory performance or the ory deficits (Leskin & White, 2007; Vasterling learning upon which it depends. et al., 1998), this is, to our knowledge, the first Additional cognitive processes that are often study to have investigated the possible association impaired in PTSD and that may relate to the between PTSD and PM. As hypothesized, the present findings include attention, working mem- results revealed a negative correlation between ory, and inhibition. Although attentional pro- PTSD symptom severity and PM performance on cesses may contribute to the relation between an event-based PM task. Specifically, veterans with PTSD and PM, basic attention, as indicated by greater PTSD symptom severity completed signifi- performance on CVLT–II Trial 1, was not signif- cantly fewer PM actions (i.e., pressed the “6” key in icantly related to PM performance in the current response to seeing “president”) than did veterans study. It is possible, however, that more complex with less severe PTSD symptoms. Our findings attentional processes not measured in the current suggest that overall PTSD symptom severity was study played a significant role. For example, associated with PM deficits and that no single working memory deficits, which have been asso- component accounted for this relationship. ciated with PTSD, could contribute to difficulty Although the strength of correlation of each of monitoring the environment for a PM cue, thus the three DSM–IV PTSD symptom clusters (i.e., reducing the likelihood of successful PM execu- reexperiencing, avoidance/numbing, hyperarousal) tion. However, this explanation is less likely in the with PM performance varied to some degree, all current study, due to the use of a focal cue. That three symptom clusters appeared to contribute to is, successful completion of the ongoing task in PM performance deficits. this study (i.e., answering trivia questions) We considered several factors that may account required “focal” processing of individual words, for these results. First, retrospective memory per- including the PM cue (“president”). Nonfocal formance has been shown to contribute to PM cues, in contrast, are cues that are not necessarily deficits (McDaniel & Einstein, 2011) and has processed in the completion of the ongoing task been found to be less proficient in PTSD (e.g., the syllable “tor”) and are therefore typically (Brewin, 2011; Schuitevoerder et al., 2013). In the more difficult to detect. Inhibition is another fac- current study, retrospective memory was not sig- tor that could have contributed to the relation Downloaded by [University of Montana] at 10:20 08 January 2016 nificantly associated with PM, suggesting that between PTSD and PM. Reduced inhibition abil- PTSD is associated with poorer PM performance, ities could have contributed to difficulty disenga- independent of retrospective memory. We also ging from the ongoing trivia task to perform the considered learning, given meta-analytic findings PM task. Because we did not examine working of impaired learning in the context of PTSD (Scott memory or inhibition in the current study, these et al., 2015). However, the pattern of results are possibilities that will need to be investigated remained unchanged when we substituted learning in future studies. performance for delayed recall in the primary We also considered alternate explanations non- regression, suggesting that differences in learning specific to PTSD. First, there was no significant ability did not account for the relation between relationship between ongoing task performance PTSD and PM. The failure to detect a statistically and PM performance, suggesting that PM perfor- significant relationship between PM and retrospec- mance differences across PTSD symptom levels tive memory and between PM and learning in this were not a product of lower task engagement in study is not surprising given the minimal demands participants with more severe PTSD. Second, we placed on retrospective memory and learning (i.e., examined the relationship between depressive there was only a single association that needed to symptoms, which frequently co-occur with PTSD 258 C. P. McFARLAND ET AL.

symptoms (Campbell et al., 2007) and have been PTSD Scale (CAPS; Blanchard et al., 1996), the shown to be associated with PM performance def- “gold standard” for diagnosing PTSD. The rela- icits under some conditions. Findings indicated tively small sample size necessitates replication that depression symptom severity was negatively with larger samples. The lack of demographic correlated with PM performance, though to a les- diversity among participants may also limit the ser degree than were PTSD symptoms. However, generalizability of findings. Additionally, although the high degree of collinearity between PTSD and we used a classic laboratory-based PM task to depression symptoms limited our ability to further determine whether a relationship exists between investigate the relationship between depression PTSD and PM, future studies would benefit from and PM. To better understand the independent inclusion of complementary ecological measures contributions of PTSD and depression symptoms of PM. on PM, we will need to replicate these findings in a Finally, we limited assessment of key potential larger sample. cognitive mechanisms subserving PM to retro- The current results are consistent with a devel- spective memory performance and measured oping picture of PM deficits in the context of only event-based PM. Based on previous research, anxiety states and behaviors (Cuttler & Graf, it is possible that the PM deficit observed in the 2007, 2008, 2009; Harris & Cumming, 2003) and current study was driven by deficits in working depressive symptomatology (Li et al., 2014; Rude memory and or inhibition (Rose, Rendell, et al., 1999), independent of retrospective memory. McDaniel, Aberle, & Kliegel, 2010; Schnitzspahn The relatively limited findings in these areas sug- et al., 2013), which might have resulted in diffi- gest that increasing levels of anxiety and depressive culty maintaining the PM intention or disenga- symptoms are related to poorer PM performance, ging from the ongoing task to initiate the consistent with the current pattern of results. intention (Kliegel, Martin, McDaniel, & Einstein, Therefore, the results of this study contribute to 2002). Examination of the relationship of PTSD our relatively nascent appreciation of the associa- to PM on PM tasks placing greater demands on tions of mood and anxiety with PM and build executive functioning, including event-based tasks upon our understanding of the psychological and that incorporate nonfocal cues or time-based tasks functional correlates of PTSD. of PM, may also be informative. It could be A wide range of daily tasks depend upon PM hypothesized, for example, that individuals with (e.g., remembering to take one’s medication and PTSD will have even greater difficulty on such completing job-related activities). Recent research PM tasks, as nonfocal cues and time-based tasks has shown that PTSD is related to functional require more self-initiated processing than the impairment in the areas of social, interpersonal, relatively simple focal event-based task employed and occupational functioning, among others in the current study. For example, individuals (Hoge et al., 2007; Rodriguez et al., 2012). It may with PTSD may have considerable difficulty enga- be that diminished PM contributes to the func- ging in the strategic monitoring of the environ- Downloaded by [University of Montana] at 10:20 08 January 2016 tional impairment associated with PTSD. ment and time that often predicts PM Similarly, PM difficulties experienced by patients performance. with PTSD may contribute to difficulties fully Limitations notwithstanding, the results of the adhering to certain aspects of both psychopharma- current study provide the first evidence of an cological (e.g., forgetting to take medications on association between PTSD symptomatology and schedule) and psychosocial (e.g., forgetting to initi- PM, adding a unique element to our understand- ate homework assignments) interventions. ing of the neurocognitive profile of PTSD. Given Therefore, better understanding of the relations the significance of PM to functioning (Kliegel, among PTSD symptoms, PM, and functional Jager, Altgassen, & Shum, 2008;Woodsetal., impairment could inform efforts to improve the 2008) combined with the significant functional level of functioning and overall quality of life of impairment in PTSD, future work would benefit many veterans. from examination of relations among PTSD, PM, Limitations of the current study include reli- and functional impairment. Progress in these ance upon the PCL-C as an indicator of PTSD realms could lead to the development of effective severity. However, there is strong agreement PM interventions tailored to the unique needs of between the PCL and the Clinician-Administered patients with PTSD. JOURNAL OF CLINICAL AND EXPERIMENTAL NEUROPSYCHOLOGY 259

Acknowledgements Cuttler, C., & Graf, P. (2009). Sub-clinical compulsive checkers show impaired performance on habitual, The authors would like to thank Sohyun Han and event- and time-cued episodic prospective memory Fabiana Cabral for their efforts in scheduling and testing tasks. Journal of Anxiety Disorders, 23, 813–823. participants. The authors report no conflicts of interest. Delis, D. C., Kramer, J., Kaplan, E., & Ober, B. A. (2000). The California Verbal Learning Test (CVLT)–Second Edition. San Antonio, TX: The Disclosure statement Psychological Corporation. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. No potential conflict of interest was reported by the W. (2002). Structured Clinical Interview for DSM-IV- author(s). TR Axis I Disorders, Research Version, Non-Patient Edition (SCID-I/NP). New York, NY: New York State References Psychiatric Institute, Biometrics Research. Forster, K. I., & Forster, J. C. (2003). DMDX: A Adda, C. C., Castro, L. H. M., Além-Mar e Silva, L. C., de Windows display program with millisecond accuracy. Behavior Research Methods, Instruments, & Manreza, M. L. G., & Kashiar, R. (2008). Prospective – memory and medial temporal epilepsy associated Computers, 35, 116 124. with hippocampal sclerosis. Neuropsychologia, 46, Geuze, E., Vermetten, E., de Kloet, C. S., Hijman, R., & 1954–1964. Westenberg, H. G. M. (2009). Neuropsychological Altgassen, M., Henry, J. D., Bürgler, & Kliegel, M. (2011). performance is related to current social and occupational functioning in veterans with posttraumatic The influence of emotional target cues on prospective – memory performance in depression. Journal of Clinical stress disorder. Depression and Anxiety, 26,7 15. and Experimental Neuropsychology, 33(8), 910–916. Gilbertson, M. W., Gurvits, T. V., Lasko, N. B., Orr, S. American Psychiatric Association. (1994). Diagnostic P., & Pitman, R. K. (2001). Multivariate assessment of and statistical manual of mental disorders (4th ed.). explicit memory function in combat veterans with Washington, DC: Author. posttraumatic stress disorder. Journal of Traumatic – Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual Stress, 14(2), 413 432. for the Beck Depression Inventory–II. San Antonio, Glisky, E. L. (1996). Prospective memory and the frontal TX: Psychological Corporation. lobes. In M. Brandimonte, G. O. Einstein, & M. A. Blake, D. D., Weathers, F. W., Nagy, L. M., Kaloupek, D. McDaniel (Eds.), Prospective memory: Theory and – G., Gusman, D. S., Charney, D. S., . . . Keane, T. M. applications (pp. 249 266). Mahwah, NJ: Erlbaum. (1995). The development of a clinician-administered Harris, L. M., & Cumming, S. R. (2003). An examina- PTSD scale. Journal of Traumatic Stress, 8(1), 75–90. tion of the relationship between anxiety and perfor- Blanchard, E. B., Jones-Alexander, J., Buckley, T. C., & mance on prospective and retrospective memory – Forneris, C. A. (1996). Psychometric properties of the tasks. Australian Journal of Psychology, 55(1), 51 55. Harris, L. M., & Menzies, R. G. (1999). Mood and PTSD checklist (PCL). Behavior Research and – Therapy, 34, 669–673. prospective memory. Memory, 7(1), 117 127. Brewin, C. R. (2011). The nature and significance of Harris, L. M., Vaccaro, L., Jones, M. K., & Boots, G. M. memory disturbance in posttraumatic stress disorder. (2010). Evidence of impaired event-based prospective – memory in clinical obsessive compulsive checking. Annual Review of Clinical Psychology, 7, 203 227. – Burgess, P. W., Gonen-Yaacovi, G., & Volle, E. (2011). Behaviour Change, 27,84 92. Functional neuroimaging studies of prospective Hoge, C. W., Terhakopian, A., Castro, C. A., Messer, S. Downloaded by [University of Montana] at 10:20 08 January 2016 memory: What have we learnt so far? C., & Engel, C. C. (2007). Association of posttrau- Neuropsychologia, 49, 2246–2257. matic stress disorder with somatic symptoms, health care visits, and absenteeism among Iraq war veterans. Burgess, P. W., Quayle, A., & Frith, C. D. (2001). Brain – regions involved in prospective memory as deter- American Journal of Psychiatry, 164, 150 153. mined by positron emission tomography. Jelinek, L., Jacobsen, D., Kellner, M., Larbig, F., Biesold, Neuropsychologia, 39, 545–555. K.-H., Barre, K., & Moritz, S. (2006). Verbal and Campbell, D. G., Felker, B. L., Liu, C.-F., Yano, E. M., nonverbal memory functioning in posttraumatic stress disorder (PTSD). Journal of Clinical and Kirchner, J. E., Chan, D., . . . Chaney, E. F. (2007). – Prevalence of depression-PTSD comorbidity: Experimental Neuropsychology, 28, 940 948. Implications for clinical practice guidelines and pri- Jenkins, M. A., Langlais, P. J., Delis, D., & Cohen, R. A. mary care-based interventions. Journal of General (2000). Attentional dysfunction associated with post- – traumatic stress disorder among rape survivors. The Internal Medicine, 22, 711 718. – Cuttler, C., & Graf, P. (2007). Sub-clinical compulsive Clinical Neuropsychologist, 14(1), 7 12. checkers’ prospective memory is impaired. Journal of Johnsen, G. E., & Asbjørnsen, A. E. (2008). Consistent – impaired verbal memory in PTSD: A meta-analysis. Anxiety Disorders, 21, 338 352. – Cuttler, C., & Graf, P. (2008). Sub-clinical checking com- Journal of Affective Disorders, 111,74 82. pulsions are related to impaired prospective memory Karstoft, K.-I., Andersen, S. B., Bertelsen, M., & Madsen, independently of depression, anxiety and distractibil- T. (2014). Diagnostic accuracy of the Posttraumatic – ity. Journal of Anxiety Disorders, 22(4), 642−654. Stress Disorder Checklist Civilian Version in a 260 C. P. McFARLAND ET AL.

representative military sample. Psychological The roles of working memory, vigilance, task regularity, Assessment, 26(1), 321–325. and cue focality. Psychology and Aging, 25(3), 595–605. Kelemen, W. L., Weinberg, W. B., Alford, H. S., Mulvey, E. Rude,S.S.,Hertel,P.T.,Jarrold,W.,Covich,J.,&Hedlund, K., & Kaeochinda, K. F. (2006). Improving the reliability S. (1999). Depression-related impairments in prospective of event-based laboratory tests of prospective memory. memory. Cognition and Emotion, 13(3), 267–276. Psychonomic Bulletin & Review, 13(6), 1028–1032. Schnitzspahn, K. M., Stahl, C., Zeintl, M., Kaller, C. P., Kliegel, M., Jager, T., Altgassen, M., & Shum, D. (2008). & Kliegel, M. (2013). The role of shifting, updating, Clinical neuropsychology of prospective memory. In and inhibition in prospective memory performance in M. Kliegel, M. A. McDaniel, & G. O. Einstein (Eds.), young and older adults. Developmental Psychology, Prospective memory: Cognitive, neuroscience, develop- 49, 1544–1553. mental, and applied perspectives (pp. 283–302). New Schnurr,P.P.,Lunney,C.A.,Bovin,M.J.,&Marx,B.P. York, NY: Erlbaum. (2009). Posttraumatic stress disorder and quality of life: Kliegel, M., Martin, M., McDaniel, M. A., & Einstein, G. Extension of findings to veterans of the wars in Iraq and O. (2002). Complex prospective memory and execu- Afghanistan. Clinical Psychology Review, 29(8), 727–735. tive control of working memory: A process model. Schuitevoerder, S., Rosen, J. W., Twamley, E. W., Ayers, Psychologische Beitrage, 44(2), 303–318. C. R., Sones, H., Lohr, J. B., . . . Thorp, S. R. (2013). A Leskin, L. P., & White, P. M. (2007). Attentional networks meta-analysis of cognitive functioning in older adults reveal executive function deficits in posttraumatic with PTSD. Journal of Anxiety Disorders, 27, 550–558. stress disorder. Neuropsychology, 21(3), 275–284. Scott, J. C., Matt, G. E., Wrocklage, K. M., Crnich, C., Li, Y. R., Loft, S., Weinborn, M., & Maybery, M. T. (2014). Jordan, J., Southwick, S. M., . . . Schweinsburg, B. C. Event-based prospective memory deficits in individuals (2015). A quantitative meta-analysis of neurocogni- with high depressive symptomatology: Problems con- tive functioning in posttraumatic stress disorder. trolling attentional resources? Journal of Clinical and Psychological Bulletin, 141(1), 105–140. Experimental Neuropsychology, 36(6), 577–587. Shucard, J. L., McCabe, D. C., & Szymanski, H. (2008). Loft, S., Doyle, K. L., Naar-King, S., Outlaw, A. Y., An event-related potential study of attention deficits Nichols, S. L, Weber, E., . . . Woods, S. P. (2014). in posttraumatic stress disorder during auditory and Allowing brief delays in responding improves event- visual go/nogo continuous performance tasks. based prospective memory for young adults living Biological Psychology, 79, 223–233. with HIV disease. Journal of Clinical and Smits, C. H. M., Deeg, D. J. H., & Jonker, C. (1997). Experimental Neuropsychology, 36(7), 761–772. Cognitive and emotional predictors of disablement in Marsh, R. L., Brewer, G. A., Jameson, J. P., Cook, G. I., older adults. Journal of Aging and Health, 9(2), 204–221. Amir, N., & Hicks, J. L. (2009). Threat-related proces- Swick, D., Honzel, N., Larsen, J., Ashley, V., & Justus, T. sing supports prospective memory retrieval for people (2012). Impaired response inhibition in veterans with with obsessive tendencies. Memory, 17, 679–686. post-traumatic stress disorder and mild traumatic Martin, M., Kliegel, M., & McDaniel, M. (2003). The brain injury. Journal of the International involvement of executive functions in prospective Neuropsychological Society, 18, 917–926. memory performance of adults. International Vasterling, J. J., Brailey, K., Constans, J. L., & Sutker, P. B. Journal of Psychology, 38, 195–206. (1998). Attention and memory dysfunction in posttrau- McDaniel, M. A., & Einstein, G. O. (2011). The neu- matic stress disorder. Neuropsychology, 12,125–133. ropsychology of prospective memory in normal Vasterling,J.J.,Duke,L.M.,Brailey,K.,Constans,J. aging: A componential approach. Neuropsychologia, I.,Allain,A.N.,&Sutker,P.B.(2002). Attention, 49, 2147–2155. learning, and memory performances and intellec-

Downloaded by [University of Montana] at 10:20 08 January 2016 McDaniel,M.A.,Glisky,E.L.,Rubin,S.R.,Guynn,M.J.,& tual resources in Vietnam veterans: PTSD and no Routhieaux,B.C.(1999). Prospective memory: A neu- disorder comparisons. Neuropsychology, 16,5–14. ropsychological study. Neuropsychology, 13, 103–110. Vedhara, K., Wadsworth, E., Norman, P., Searle, A., Racsmány, M., Demeter, G., Csigó, K., Harsányi, A., & Mitchell, J., Macrae, N., . . . Memel, D. (2004). Németh, A. (2011). An experimental study of Habitual prospective memory in elderly patients with prospective memory in obsessive-compulsive disor- Type 2 diabetes: Implications for medication adher- der. Journal of Clinical and Experimental ence. Psychology, Health, & Medicine, 9(1), 17–27. Neuropsychology, 33(1), 85–91. Weathers, F. W., Litz, B. T., Herman, D. S., Huska, J. A., & Rauch, S. L., Shin, L. M., & Phelps, E. A. (2006). Keane, T. M. (1993, October). The PTSD Checklist Neurocircuitry models of posttraumatic stress disorder (PCL): Reliability, validity, and diagnostic utility. Paper and extinction: Human neuroimaging research—past, presented at the Annual Meeting of the International present, and future. Biological Psychiatry, 60, 376–382. Society for Traumatic Stress Studies, San Antonio, TX. Rodriguez, P., Holowka, D. W., & Marx, B. P. (2012). Wechsler, D. (2001). Wechsler Test of Adult Reading.San Assessment of posttraumatic stress disorder-related func- Antonio, TX: The Psychological Corporation. tional impairment: A review. Journal of Rehabilitation Woods,S.P.,Iudicello,J.E.,Moran,L.M.,Carey,C. Research and Development, 49(5), 649–666. L.,Dawson,M.S.,&Grant,I.(2008). HIV-asso- Rose, N. S., Rendell, P. G., McDaniel, M. A., Aberle, I., & ciated prospective memory impairment increases Kliegel, M. (2010). Age and individual differences risk of dependence in everyday functioning. in prospective memory during a “Virtual Week”: Neuropsychology, 22(1), 110–117.