Contributions of Personality and Depressive Symptoms on Subjective and Objective Memory

Abstract

This study aimed to examine whether individual differences in personality and multifaceted depression explain discrepancies between subjective memory (SM) and objective memory (OM), and whether these relationships varied as a function of gender. Structural equation modeling was used to test these relationships in a group of older adults (65–98 years, N = 606) from the 2000 Wave Panel 3 of the Long Beach Longitudinal Study. Women outperformed men on OM, reporting less frequent memory failures. Dysphoria and openness predicted OM, yet not SM. Cognitive concerns, neuroticism, and conscientiousness predicted SM but not OM. The personality and Geriatric Depression Scale (GDS) factors relevant for SM differed from those for OM, with the GDS factors showing a stronger impact. Overall, discrepancies between SM and OM may be partially explained by the varying impact of the GDS and personality factors, as both provide differential utility in understanding SM and OM differences in older adulthood.

Keywords

subjective memory objective memory aging depressive symptoms personality

Introduction

Assessment of subjective memory (SM) could be a useful indicator of pathological cognitive change if individuals accurately estimated their memory function (i.e., objective memory [OM]; Slavin et al., 2010). The relationship between SM and OM in the literature has been inconclusive, suggesting either no association between SM and OM performance (Jungwirth et al., 2004; O’Connor, Pollitt, Roth, Brook, & Reiss, 1990; Pearman, Hertzog, & Gerstorf, 2014; Pearman & Storandt, 2005) or only a moderate relationship (Gilewski, Zelinski, & Schaie, 1990; Jonker, Gerrlings, & Schmand, 2000; Kahn, Zarit, Hilbert, & Niederehe, 1975; Lam, Lui, Tam, & Chiu, 2005; Lane & Zelinski, 2003; Pearman et al., 2014; Pearman & Storandt, 2004; Zelinski, Burnight, & Lane, 2001). Kahn et al. (1975) first reported dissociation between SM and OM, suggesting that SM showed more of an association with depressive symptoms than with actual OM performance.

Since this initial study, the impact of individual differences in psychological factors (e.g., depressive symptoms and personality factors) on SM and OM has been questioned by researchers (Hülür, Hertzog, Pearman, Ram, & Gerstorf, 2014; Pearman et al., 2014). Many studies have demonstrated that individuals with higher levels of depressive symptoms reported lower scores on their SM (Hülür et al., 2014; Kahn et al., 1975; Pearman et al., 2014). Depressive symptoms can negatively influence self-perceptions of memory, as people who are frustrated or under stress tend to have pessimistic views (Hülür et al., 2014). Depressive symptoms have also been negatively associated with OM since individuals with depressive symptoms may experience dysfunction in concentration (Steffens & Potter, 2008)

Regarding personality, both differential and similar relationships have been observed with SM and OM. SM has been negatively related to neuroticism (Lane & Zelinski, 2003; Merema, Speelman, Foster, & Kaczmarek, 2013; Neupert, Mroczek, & Spiro, 2008; Pearman & Storandt, 2005; Slavin et al., 2010; Zelinski & Gilewski, 2004) and positively related to conscientiousness (Pearman & Storandt, 2005; Slavin et al., 2010; Zelinski & Gilewski, 2004) and openness (Slavin et al., 2010). Previous studies demonstrated OM correlated with personality traits, reporting significant relationships between personality and OM performance (Boyle et al., 2010; Graham & Lachman, 2012; Gregory, Nettelbeck, & Wilson, 2010; Meier, Perrig-Chiello, & Perrig, 2002; Soubelet & Salthouse, 2011; Wilson et al., 2003, 2007). OM has been positively associated with openness (Graham & Lachman, 2012; Gregory et al., 2010; Schaie, Willis, & Caskie, 2004), extraversion (Meier et al., 2002, Pearman, 2009; Schaie et al., 2004), and conscientiousness (Baker & Bichsel, 2006), while it has been negatively associated with neuroticism (Graham & Lachman, 2012; Wilson et al., 2003, 2007). Of the personality traits, overall findings suggested that conscientiousness and openness were positively related to SM and OM, whereas neuroticism was negatively associated with both types of memory.

For both SM and OM, findings across studies have concluded inconsistent results, reporting different magnitudes of associations with psychological factors. These incongruent findings may be accounted for by methodological differences. Different studies used samples with variations in age, education, race, and other inclusion criteria (Boyle et al., 2010; Soubelet & Salthouse, 2011). Some studies included participants with dementia (Boyle et al., 2010; Pocnet, Rossier, Antonietti, & von Gunten, 2013; Terry, Puente, Brown, Faraco, & Miller, 2013), while other studies excluded individuals based on the degree of cognitive impairment (Allen, Kaut, Baena, Lien, & Ruthruff, 2011; Cuttler & Graf, 2007; Gregory et al., 2010; Pearman & Storandt, 2004; Soubelet & Salthouse, 2011). Furthermore, depression was considered an exclusionary criterion in some studies (Pearman & Storandt, 2004; Pocnet et al., 2013), while participants with depression were included in other studies (Ayotte, Potter, Williams, Steffens, & Bosworth, 2009; Boyle et al., 2010; Terry et al., 2013). Another possible reason for differential results in prior research may be that depressive symptoms were analyzed as a single score of depressive symptoms (i.e., depressive symptoms are summed up as a single score). If depression was measured as a single manifest variable (i.e., total scores of depressive symptoms), then underlying latent factors were not assessed as individual predictors (Haavisto & Boron, 2018). A few early studies examined interrelationships between factors of the Geriatric Depression Scale (GDS; Yesavage et al., 1983) and OM performance (Bentz & Hall, 2008; Havins, Massman, & Doody, 2012), yet these studies did not examine relationships of SM with OM and the GDS subfactors.

Most recently, Haavisto and Boron (2018) detected a four-factor solution (dysphoria, vigor or withdrawal, cognitive concerns, and agitation) of the GDS by using exploratory factor analyses. A confirmatory factor analysis was performed to examine the construct validation of the four-factor model in a sample of 606 healthy community-dwelling adults over the age of 65 years. This factor solution was employed to examine whether the multifaceted depressive symptoms explained the differences between SM and OM by controlling age and educational attainment. Results indicated that dysphoria negatively predicted OM, yet not SM, and cognitive concerns negatively predicted SM, but not OM, suggesting that such relationships cannot be discovered when assessing depression as a single score (i.e., the GDS total scores). Yet, contributions of personality traits on SM and OM while assessing depression as multifaceted latent factors were not considered; the potential overlap between depressive symptoms and personality (Barnhoper & Chittka, 2010; Merema et al., 2013) may remain in the association between SM and OM. Merema et al. (2013) demonstrated that the association between depressive symptoms and SM diminished among healthy community-dwelling older adults (66–90 years; N = 177) in Western Australia after variance associated with age, gender, education, OM, conscientiousness, and neuroticism was partialled out. Yet, Merema et al. did not test the nature of multifaceted depressive symptoms; thus, the impact of personality traits and depressive symptoms on SM and OM should be further explored.

Furthermore, Haavisto and Boron (2018) did not test the impact of gender in their study on the relationships between the GDS subfactors and SM and OM. Previous studies demonstrated that gender was differentially related to personality traits (Chapman, Duberstein, Sörensen, & Lyness, 2007), depressive symptoms (Zunzunegui et al., 2007), and OM (Gerstorf, Herlitz, & Smith, 2006). Chapman et al. examined gender difference in a sample of 486 older adults (age range: 65–98 years; mean age = 75 years; SD = 6.5), using the NEO-Five Factor Inventory (Costa & McCrae,1992b). They found higher levels of neuroticism and agreeableness in women than men. Zunzunegui et al. (2007) investigated country-specific gender differences in depressive symptoms with older men and women from five countries (i.e., Italy, The Netherlands, Spain, Sweden, and Israel), suggesting that depressive symptoms in women exceeded those in men in every country, except Sweden. Gerstorf et al. (2006) examined patterns of gender differences on cognitive change during advanced old age with 13-year longitudinal data from the Berlin Aging Study (N = 368; age range: 70–100 years; mean age = 83 years at baseline assessment). They found women outperformed men on tasks to assess episodic memory while controlling for gender- and cohort-related differences in education. Given this prior work, the primary aims of this study were to examine the impact of personality on the relationship between depression and memory (SM and OM), and whether these relationships varied as a function of gender. The four GDS factor model (dysphoria, vigor or withdrawal, cognitive concerns, and agitation; Haavisto & Boron, 2018) was employed to investigate the differential impact of individual variations in personality and depressive symptoms on SM and OM because the four factors were derived from the same sample as this study (Figure 1). As shown in Figure 1, the impact of each personality factor on SM and OM can be simultaneously tested while teasing out the impact of depressive symptoms on SM and OM, controlling for age, gender, and educational attainment.

Figure 1.

Proposed structural model depicted relationships between age, gender, education, personality traits, depressive symptoms, SM, and OM. GFF = General Frequency of Forgetting; SF = Seriousness of Forgetting; RF = Retrospective Functioning; MU=Mnemonic Usage.

Method

Sample and Procedure

The sample was taken from the 2000 Wave Panel 3 of the Long Beach Longitudinal Study (LBLS; Zelinski & Kennison, 2011). Panel 3 was recruited by either sampling from membership lists of the Southern California Health Maintenance Organization, Family Health Plan, or recruiting with age-targeted direct mail and advertisements in senior newspapers. LBLS testers administered the Mini-Mental State Examination (Folstein, Folstein, & McHugh, 1975) to screen out people who showed mild cognitive impairment (i.e., those who scored less than 25 on the Mini-Mental State Examination). All participants were healthy community-dwelling adults. The 916 individuals of Panel 3 were first tested in 2000 to 2001. Participant ages ranged from 29 to 98 years and nearly a third of the Panel 3 participants were over the age of 80 years (Lewis & Zelinski, 2010; Zelinski, Kennison, Watts, & Lewis, 2009).

Participants

Of the 2000 Wave Panel 3 from LBLS, only participants who were older than 64 years (N = 606; age range: 65–98; M = 77.15; SD = 7.03) were used in this study, as the purpose was focused on how individual differences in personality traits and depressive symptoms can explain the inconsistency between SM and OM among older adults. Subjects had between 4 and 20 years of education (M = 14.43; SD = 2.68) and 52.1% of them were female. Table 1 presents means, standard deviations, and correlations of all of the selected variables in the proposed model. Based on the cutoff values of the GDS total scores, 92% of participants were in the normal range (0–10). Of the sample, 7.3% showed mild depressive symptoms (11–20) and 0.8% showed severe depressive symptoms (21–30). A one-way analysis of covariance was conducted to test whether depressive symptoms (i.e., mild to severe depressive symptoms) disturbed OM performance. Equality of variances for the compared groups was examined by using Levene’s Test. No mean difference in OM performance between the groups was found while controlling for age, gender, and education.

Table 1.

Means, Standard Deviations, and Zero Correlations of all of the Variables in the Proposed Model.

	Mean (SD)	Age	Gender	Education	N	E	O	A	C	GDS	GFF	SF	RF	MU	IR I	IR II	TR I	TRII
Gender	1.52 (.50)
Education	14.51 (2.65)	−.16**	−.18**
N	1.58 (.41)	.03	.13**	−.18**
E	2.17 (.36)	−.19**	−.03	.01	−.27**
O	2.21 (.35)	−.23**	.11*	.22**	−.10**	.34**
A	2.66 (.33)	.10*	.19**	−.06	−.23**	.08	.10*
C	2.53 (.37)	−.08	−.02	.12*	−.49**	.37**	.10*	.18**
GDS	4.28 (4.37)	.08	.10*	−.04	.52**	−.32**	−.13**	−.11*	−.34**
GFF	4.84 (.88)	−.13*	.16**	.05	−.32**	.22**	.15**	.18**	.38**	−.35**
SF	4.67 (1.21)	.03	.10*	.01	−.22**	.10**	.03	.08	.20**	−.18**	.42**
RF	3.55 (1.08)	.02	.11*	−.12*	−.02	.09	.02	.02	.08	−.12*	.20**	.11*
MU	3.16 (1.37)	−.08	−.14**	−.03	−.09	−.03	−.11*	.01	−.13**	.01	.09	.15**	.04
LR I	.56 (.19)	−.29**	.14**	.21**	−.10*	.05	.18**	.10*	.07	−.16**	.25**	.05	−.10*	−.02
LR II	.45 (.20)	−.25**	.15**	.23**	−.07	.03	.18**	.08	.01	−.08	.22**	.11*	−.09	.02	.65**
TR I	.24 (.12)	−.28**	.09	.24**	−.07	.01	.17**	.00	−.03	−.08	.14**	−.03	−.08	.02	.39**	.41**
TRII	.22 (.12)	−.30**	.10*	.25**	−.06	.02	.20**	−.01	.00	−.07	.13**	−.03	−.08	.02	.43**	.41**	.72**
TRIII	.30 (.15)	−.25**	−.13**	.34**	−.10*	−.02	.17**	−.04	−.05	−.09	.12**	−.06	−.11*	.05	.40**	.35**	.69**	.71**

Note. N = neuroticism; E= extraversion; O = openness; A =agreeableness; C = conscientiousness; GDS total scores = Geriatric Depression Scale total scores; GFF = General Frequency of Forgetting; LR = list recall; TR = text recall; SF = Seriousness of Forgetting; RF = Retrospective Functioning; MU = Mnemonic Usage.

**p < .01; *p < .05.

Memory-Related Physical Health Conditions

Previous studies indicated physical health conditions (e.g., hypertension and diabetes) were associated with OM performance (Okereke et al., 2008; Saxby, Harrington, McKeith, Wesnes, & Ford, 2003). Participants were asked about their chronic physical health conditions and psychiatric disorders (e.g., “Heart problems?”, “Needs help walking?”, and “Psych problems?”); yet those self-reported assessments were not used as exclusion criteria. A series of one-way analyses of covariance was conducted when controlling for age, gender, education, depression (assessed by the GDS total scores), personality traits (assessed by the five personality mean scores), and whether those participants’ health conditions (i.e., diabetes, heart problems, hypertension, psychiatric problems, and needing help walking) may result in distorted findings in testing the proposed structural model. Levene’s F tests were conducted to investigate equality of variance while comparing the groups (e.g., hypertensive group vs. nonhypertensive group). Individuals who endorsed always needing assistance with walking showed slightly lower scores on Immediate Recall I (p < .01; partial eta squared = .021) compared with those without need for constant help with walking. People with high blood pressure showed slightly lower scores in Text Recall (TR) I (p< .05; partial eta squared = .008) than those without high blood pressure. Overall, the effects of those health conditions on OM performance were negligible. Those health conditions were not considered as confounding factors in this study.

Measures

Subjective memory

The Memory Function Questionnaire (MFQ) was used to measure SM (Gilewski & Zelinski, 1988). It is composed of four factors (General Frequency of Forgetting [GFF], Seriousness of Forgetting [SF], Retrospective Functioning [RF], and Mnemonic Usage [MU]). Higher scores in each domain indicate higher levels of SM performance. The MFQ has shown high test–retest reliability and concurrent validity (Gilewski & Zelinski, 1988; Gilewski et al., 1990; Zelinski, Gilewski, & Anthony-Bergstone, 1990). Four mean scores were computed for the four MFQ factors as shown in Figure 1. The four factors of the MFQ were analyzed as manifest variables.

Objective memory

List recall (LR) was measured by two subtests: Immediate Recall 1 and 2. Participants studied a list of 20 high-frequency words for 3½ minutes (Zelinski & Lewis, 2003). Once the studying time was completed, they were then asked to immediately recall as many of the words as possible. The amount of time to recall them was unlimited. The recall lists for Immediate Recall 1 and 2 were formed in a similar pattern and were administered during different testing sessions Each word on the list was scored as 1 (correctly recalled) or 0 (not recalled) (Lewis & Zelinski, 2010).

TR was composed of three separate subtests: TR 1 (a 227-word essay), Text 2 (a 204-word essay), and Text 3 (a 209-word essay). Participants were required to hear three separate essays read loud (at 155 words per minute) and then immediately free recall as much of the essay as possible by writing (Lewis & Zelinski, 2010). The score of each TR test was computed as the proportion of correctly recalled propositions. Each word on the text was scored as 1 (correctly recalled) or 0 (missed) (Zelinski & Lewis, 2003).

To test the validity of OM tests (LR and TR), a confirmatory factor analysis was conducted. Mean scores (range: 0–1) were created for each OM test (Immediate Recall I and II and TR I, II, and III) to assess the OM factor structure. The cutoff values for excellent fit indices were root mean square of error approximation (RMSEA) ≤.05, comparative fit index (CFI) ≥.95, Tucker–Lewis Index (TLI) ≥ .95, while the cutoff values for acceptable fit indices were RMSEA ≤.08, CFI ≥ .90, and TLI ≥ .90 (Brown, 2014). The model fit indices yielded: χ² (4, n = 603) = 8.545, p =.074; RMSEA = .043; CFI = .997; TLI = .992; and SRMR = 0.009, indicating an excellent model fit. This OM measurement portion is included in the proposed structural model as seen in Figure 1.

Personality

In 2000, participants completed the Revised NEO Personality Inventory (Costa & McCrae, 1992a). The Revised NEO Personality Inventory includes five personality domains to assess traits of neuroticism, extraversion, openness, agreeableness, and conscientiousness. Internal consistency for neuroticism, extraversion, openness, agreeableness, and conscientiousness was reported as high, with alphas of .93, .87, .89, .76, and .86, respectively (Costa & McCrae, 1992a). Five mean scores were created for neuroticism, extraversion, openness, agreeableness, and conscientiousness, and they were analyzed as manifest variables.

The GDS Factor Structure

The GDS (Yesavage et al., 1983) was used to assess depressive symptoms for participants in the LBLS. The scale comprises 30 items and the scores on each item ranged from 0 (no) and 1 (yes). Of the 30 items in the GDS, 20 items indicate the presence of depressive symptoms (e.g., “Do you feel that your life is empty?”) when answered positively (yes). In contrast, 10 other items (Question Numbers 1, 5, 7, 9, 15, 19, 21, 27, 29, and 30) indicate depression (e.g., “Are you hopeful about the future?”) when answered negatively (no). For this reason, the responses of the 10 items (Question Numbers 1, 5, 7, 9, 15, 19, 21, 27, 29, and 30) were recoded in order to establish a consistent direction with their labels. Usually, the total score of the GDS is summed as 1 point for each of the negative answers (indicating depressive symptoms). The cutoff values are normal (0–10), mild depressive symptoms (11–20), and severe depressive symptoms (21–30). However, the depressive symptoms of the participants were not clinically diagnosed, and the cutoff values of the GDS were not used for testing the proposed model (Figure 1) in this study.

Haavisto and Boron (2018) obtained the four-factor solution of the GDS by testing the same sample used for this study. Exploratory factor analyses were conducted to seek latent factors of the GDS with geomin rotation. Results showed the 28 items fit into four factors named dysphoria, vigor/withdrawal, cognitive concerns, and anxiety. A confirmatory factor analysis was conducted to test the validity of the four-factor GDS solution, yielding good ranges of fit indices, χ²(343, n = 604) = 498.48, p <.001; RMSEA = .027; CFI = .957; and TLI = .952 (Haavisto & Boron, 2018). Accordingly, the GDS factor structure comprised the four factors (dysphoria, vigor/withdrawal, cognitive concerns, and agitation) that served as the measurement portion for the following structural equation modeling (SEM).

Statistical Analyses

SEM was performed by using Mplus 7.0 (Muthén & Muthén, 1998–2015). The WLSMV estimator was utilized to analyze dichotomous variables (i.e., the GDS items). Unimputed direct WLSMV was used to address missing data, as it would outperform list wise deletion considering the small portion of missing data in the sample (Asparouhov & Muthén, 2010). Prior to running SEM, measurement models were tested. A CFA was conducted to examine the OM measurement portion. EFAs were performed to extract the four GDS factor solution and a CFA was conducted to test the validity of the four GDS factors (Haavisto & Boron, 2018).

Results

A structural model was proposed (Figure 1) and tested by SEM as shown in Figure 2. Acceptable fit indices of χ² (df = 803, n = 552) = 1090.832, p < .001; RMSEA = .025; CFI = .941; and TLI = .928 were obtained for this model. More detailed SEM results are listed in Tables 2 and 3. Table 2 presents standardized estimates of the measurement models for the GDS (Yesavage et al., 1983) and OM. Table 3 depicts standardized regression coefficients for the outcome variables from the SEM results.

Figure 2.

SEM results: Standardized coefficients and correlations. Note that fit indices: χ²(df = 803, n = 552) = 1090.832, p < .001; RMSEA = .025; CFI = .941; TL = .928. *p < .05. **p <.01. ***p < .001. SEM results in more detail are listed in Tables 2 to 4 due to limited space in this figure. CFAs and EFAs results of the GDS factors and the CFA results of the OM factor structure will be provided upon request. GFF = General Frequency of Forgetting; SF = Seriousness of Forgetting; RF = Retrospective Functioning; MU = Mnemonic Usage

Table 2.

Standardized Estimates of Measurement Models From SEM Results.

	Measurement model		Estimates (SE)
GDS
	Dysphoria
		GDS1	.77*** (.06)
		GDS2	.65*** (.05)
		GDS3	.82*** (.06)
		GDS4	.75*** (.04)
		GDS10	.86*** (.04)
		GDS11	.67*** (.05)
		GDS16	.95*** (.03)
		GDS23	.54*** (.07)
		GDS24	.66*** (.05)
		GDS25	.76*** (.06)
	Vigor/withdrawal
		GDS5	.84*** (.04)
		GDS12	.56*** (.05)
		GDS15	.67*** (.09)
		GDS17	.77*** (.06)
		GDS19	.73*** (.04)
		GDS20	.50*** (.07)
		GDS21	.78*** (.04)
		GDS22	.83*** (.08)
		GDS27	.56*** (.06)
		GDS28	.62*** (.06)
	Cognitive concerns
		GDS14	.66*** (.06)
		GDS20	.88*** (.05)
		GDS20	.31*** (.08)
		GDS30	.71*** (.06)
	Agitation
		GDS6	.85*** (.05)
		GDS8	.77*** (.07)
		GDS13	.77*** (.05)
		GDS18	.82*** (.06)
		GDS29	.54*** (.06)
OM
	List recall
		Immediate recall I	.88*** (.03)
		Immediate recall II	.76*** (.03)
	Text recall
		Text recall I	.86*** (.02)
		Text recall II	.87*** (.02)
		Text recall III	.84*** (.02)

***p < .001.

GDS= Geriatric Depression Scale; OM = objective memory.

Table 3.

Standardized Regression Coefficients From SEM Results.

	Outcome
Predictor	GFF estimate (SE)	SF estimate (SE)	RF estimate (SE)	MU estimate (SE)	LR estimate (SE)	TR estimate (SE)
Age	−.07 (.05)	.05 (.05)	.03 (.04)	−.13** (.05)	−.33*** (.05)	−.33*** (.05)
Gender	.13** (.05)	.09 (.05)	.08 (.05)	−.17*** (.05)	.21*** (.05)	.12* (.05)
Education	.00 (.05)	.01 (.06)	−.06 (.06)	−.05 (.05)	.16** (.06)	.25*** (.05)
Neuroticism	−.08 (.05)	−.16 (.08)	.05 (.08)	−.21* (.08)	.09 (.09)	.01 (.08)
Extraversion	.11 (.10)	.06 (.11)	−.01 (12)	.01 (12)	.10 (.12)	.09 (.12)
Openness	.06 (.04)	.01 (.05)	−.01 (.05)	−.01 (.05)	.11 (06)	.10* (.05)
Agreeableness	.07 (.05)	−.02 (.05)	.05 (.05)	.06 (.05)	.08 (.05)	.02 (.05)
Conscientiousness	.14 (.06)	.02 (.06)	.00 (.07)	−.26*** (.06)	−.01 (.07)	−.11 (.06)
Dysphoria	−.19 (.20)	−.10 (.23)	.27 (.22)	.06 (.23)	−.53* (.24)	−.35 (.22)
Vigor/withdrawal	−.17 (.16)	.03 (.20)	−.08 (.17)	.04 (.18)	.26 (.20)	.29 (.19)
Cognitive concerns	−.50*** (.09)	−.38*** (.10)	−.22* (.09)	−.20 (.10)	−.07 (.10)	−.07 (.10)
Agitation	.20 (.14)	.29 (.16)	−.22 (.16)	.13 (.14)	.22 (.18)	.05 (.15)
Residual variances (SE)	.62*** (.06)	.82*** (.05)	.91*** (.05)	.90*** (.03)	.68*** (.06)	.71*** (.05)

	Outcome
Predictor	Neuroticism estimate (SE)	Extraversion estimate (SE)	Openness estimate (SE)	Agreeableness estimate (SE)	Conscientiousness estimate (SE)	Dysphoria estimate (SE)	Vigor/withdrawal estimate (SE)	Cognitive concerns estimate (SE)	Agitation estimate (SE)
Age	.04 (.05)	−.15** (.04)	−.20*** (.04)	.10* (.04)	−.05 (.05)	−.01 (.06)	.13* (.05)	.01 (.06)	−.03 (.06)
Gender	.07 (.04)	.03 (.05)	.18*** (.04)	.19*** (.04)	.03 (.05)	.12* (.06)	.13* (.05)	−.03 (.06)	.16** (.06)
Education	−.15** (.04)	−.02 (.05)	.24*** (.04)	−.02 (.04)	.10 (.05)	−.12 (.06)	.07 (.05)	−.03 (.06)	.00 (.06)
Residual variances (SE)	.97*** (.02)	.98*** (.01)	.88*** (.03)	.95*** (.02)	.99*** (.01)	.97*** (.02)	.98*** (.02)	1.0*** (.01)	.97*** (.02)

Note. GFF = General Frequency of Forgetting; SF = Seriousness of Forgetting; RF = Retrospective Functioning; MU = Mnemonic Usage; LR = list recall; TR = text recall; SE = standard error.

*p < .05; **p < .001; ***p < .001.

Personality and Depression

In the structural equation model seen in Figure 2, age had a positive effect on the GDS subscale vigor/withdrawal (β = .13, p < .05), suggesting that older participants reported higher vigor/withdrawal than younger participants. Age positively predicted agreeableness (β = .10, p < .05), while it negatively predicted extraversion (β = −.15, p < .01) and openness (β = −.20, p < .001); older participants were more agreeable than younger participants, and younger people reported being more extroverted and more open to new experiences than older people. Gender had positive effects of dysphoria (β = .12, p < .05) and agitation (β = .16, p < .01), indicating that women reported more dysphoric moods and agitation than men. Gender had positive relationships with openness (β = .18, p < .001) and agreeableness (β = .19, p < .001), suggesting that women were more open to new experiences and more agreeable than men. Education was found to have a negative effect on dysphoria (β = −.12, p < .05). Education negatively predicted neuroticism (β = −.15, p < .01), while it positively predicted openness (β = .24, p < .001), suggesting that the more education people had, the less neurotic and the more open to new experiences they were.

Table 4 depicts the correlations between personality factors and the four GDS factors as well as the association between SM and OM derived from the SEM results in Figure 2. Overall, of the personality factors, only neuroticism was positively associated with dysphoria (r = .61, p < .001), vigor/withdrawal (r = .46, p < .001), cognitive concerns (r = .46, p < .001), and agitation (r = .58, p < .001), indicating more neurotic people reported more depressive symptoms. Conscientiousness showed negative associations with dysphoria (r = −.61, p < .001), vigor/withdrawal (r = .61, p < .001), cognitive concerns (r = .61, p < .001), and agitation (r = .61, p < .001), suggesting more conscientious people endorsed fewer depressive symptoms. Of the GDS factors, vigor/withdrawal showed a significantly strong negative correlation with extroversion, while other GDS factors had weak correlations with extroversion. It suggested that more extroverted individuals reported more energetic moods and more interest in social engagement (reflecting the vigor/withdrawal factor) than less extroverted people. Openness and agreeableness showed no to negligible correlations with the GDS factors.

Table 4.

Correlations Between the Variables From SEM Results.

	N	E	O	A	C	Dysphoria	Vigor/Withdrawal	Cognitive Concerns
E	−.31***
O	−.08*	.31***
A	−.25***	.11**	.12**
C	−.50***	.39***	.06	.21***
Dysphoria	.61***	−.16**	−.08	−.17**	−.27***
Vigor/Withdrawal	.46***	−.60***	−.19***	−.11*	−.38***	.71***
Cognitive Concerns	.46***	−.25***	−.07	−.07	−.42***	.58***	.53***
Agitation	.58***	−.27***	−.14*	−.05	−.33***	.76***	.59***	.59***
	GFF	SF	RF	MU	LR
SF	.27***
RF	.10*	.06
MU	.12*	.16***	.08
LR	.17**	.04	.14*	−.02
TR	.11	−.06	−.07	.01	.46***

Note. N = neuroticism; E= extraversion; O = openness; A=agreeableness; C = conscientiousness; GFF = General Frequency of Forgetting; SF = Seriousness of Forgetting; RF = Retrospective Functioning; MU=Mnemonic Usage; LR = list recall; TR = text recall.

*p < .05; **p < .001; ***p < .001.

SM and OM

Of the MFQ (Gilewski & Zelinski, 1988) factors (i.e., SM), GFF and RF had significant, yet weak associations with LR, suggesting that people with lower performance in recalling words reported more memory problems and considered their forgetfulness more seriously. Of the GDS factors, only cognitive concerns predicted SM. Cognitive concerns was negatively associated with GFF (β = −.50, p < .001) and accounted for 25% of the variance. Other predictors of GFF were conscientiousness (β = .14, p < .05) and gender (β = .13, p < .01), suggesting that more conscientious people reported less frequent memory failures, whereas men reported more frequent memory problems than women. Cognitive concerns (β = −.38, p < .001) and neuroticism (β = −.18, p < .05) negatively predicted SF; people who were more concerned about cognitive decline considered their forgetfulness more seriously. Furthermore, people scoring higher on neuroticism considered their forgetfulness more seriously. Cognitive concerns was the only predictor of RF and was negative (β = −.22, p < .05). Age (β =−.13, p < .01), gender (β = −.17, p < .001), neuroticism (β = −.20, p < .05), and conscientiousness (β = −.26, p < .001) negatively predicted Mnemonics Usage. Older participants reported using mnemonics more often than younger people. Women relied on mnemonics more than men. In addition, those scoring higher on neuroticism or conscientiousness reported more frequent mnemonic adoption.

Age negatively predicted LR (β = −.33, p < .001) and TR (β = −.33, p < .001), suggesting that younger people performed better at recalling words and texts than older participants. Education positively predicted LR (β = .16, p < .01) and TR (β = .25, p < .001). Gender showed positive effect on LR (β = .21, p < .001) and TR (β = .12, p < .05), indicating that women scored higher on both recall tests compared with men. Of the GDS factors, dysphoria was the only predictor of LR (β = −.53, p < .05) and accounted for 28.1% of the variance. People with a more dysphoric mood scored lower on word recall compared with those who scored lower on dysphoric mood. Of the personality traits, openness positively predicted TR (β = .10, p < .05).

Discussion

This study examined the underlying relationships between personality traits, depressive symptoms, SM, and OM while controlling for age, gender, and education. A major finding of this study was that each GDS (Yesavage et al., 1983) and personality factor showed differential relationships with SM and OM, indicating that the discrepancy of SM and OM may be explained by how individual differences in personality and depressive symptoms influence SM and OM. The factor of the GDS (cognitive concerns) that influenced SM differed from the GDS factor (dysphoria) that influenced OM in this study. The findings were congruent with prior work (Haavisto & Boron, 2018). However, compared with prior research, by employing personality factors and gender in this study, the impact of vigor/withdrawal on OM was diminished, yet the impact of dysphoria on OM was strengthened. This is a novel finding and demonstrates the potential utility of including both personality and GDS factors when investigating SM and OM.

Strong correlations between neuroticism and depressive symptoms were found in previous studies (Barnhoper & Chittka, 2010; Weber et al., 2012). In this study, personality factors were employed to tease apart overlapping effects of personality and depressive symptoms on SM and OM. Interestingly, differential correlations were discovered between personality traits and the GDS latent factors. For example, extroversion showed a strong negative association with vigor/withdrawal, while it had weak to moderate relationships with other GDS factors. Thus, consideration of both personality and depressive symptoms when OM and SM are of interest adds some additional value beyond simply utilizing depression or personality factors alone.

Beyond the investigation of the impact of personality and depression on OM and SM, gender differences were also of interest. This study demonstrated consistent results with previous studies (Gerstorf et al., 2006) in that women performed better on OM tests. Gender differences were also observed in personality traits. Older women were more agreeable than older men, consistent with prior research. The gender differences found in openness, and the lack of differences in neuroticism were incongruent with prior work (Chapman et al., 2007). One possible reason for the incongruent results may be that Chapman et al. did not partial out the impact of depressive symptoms when examining gender differences in personality factors, considering the high correlations between neuroticism and depression in this study, this suggests the lack of consideration of personality when examining depression, or vice versa, may be problematic. With regard to depression, women reported more symptoms, particularly in dysphoria and agitation, than men, congruent with earlier work (Zunzunegui et al., 2007). Thus, consideration of these gender differences in depressive symptoms in the current model may help explain the gender discrepancies found in comparison to other research.

Prior research has suggested that the association between SM and depression is stronger than the relationship between SM and OM (Kahn et al., 1975; Jessen et al., 2007; Jonker, Launer, Hooijer, & Lindeboom, 1996; Minett, Da Silva, Ortiz, & Bertolucci, 2008; Zimprich, Martin, & Kliegel, 2003). Employing personality factors and the GDS factor structure, as was done in this study, might lead to more advanced findings from this early speculation. More dysphoric mood significantly disturbed recalling words, and individuals in the sample did not notice their memory problems considering there was no association between dysphoria and SM (GFF). More worries about cognitive decline did not influence one’s OM performance. Similar with the patterns of relationships for the GDS factors, SM, and OM, the personality factors (i.e., neuroticism and conscientiousness) that predicted SM differed from those (i.e., openness) that predicted OM. Yet, the GDS factors showed a stronger impact on the discrepancy between SM and OM compared with the personality factors. Overall, this differential impact of the GDS factors and personality factors on SM and OM played a critical role in discrepancies between SM and OM.

The findings of this study must be interpreted in the context of several methodological limitations. First, different measures of OM and personality may have partially contributed to the inconsistent findings between studies (Curtis, Windsor, & Soubelet, 2015; Soubelet & Salthouse, 2011). For example, Baker and Bichsel (2006) found that conscientiousness was associated with short-term memory, but not with long-term memory by using the Woodcock-Johnson Test of Cognitive Abilities. Gregory et al. (2010) found that openness was positively related to short-term memory, but not to working memory by using the Wechsler memory scale-III. Thus, this study used different measures of memory domains (i.e., LR and TR) and psychological factors (i.e., personality and depressive symptoms) from previous studies that may have partially contributed to the incongruent findings compared with prior research. The second limitation is that, due to the cross-sectional design, it is impossible to tease apart the effects of neuroticism and depressive symptoms; a longitudinal design with time-varying variables would clarify the individual contributions of neuroticism and depression on SM and OM.

Finally, due to the cross-sectional design of this study, it may be difficult to claim causal relationships as shown in the hypothesized model. Yet, Haavisto and Boron (2018) argued that age and gender were determined when an individual was born, so these can be the predictors of all other variables in the hypothesized model in Figure 1. Many studies have found that depressive symptoms are positively associated with SM and OM; however, the direction of these relationships is unknown and may be even be reciprocal. In this model, depressive symptoms were hypothesized as antecedents to SM and OM, which is a limitation. Alternatively, personality traits remain stable after people reach 30 years of age (Costa & McCrae, 1988). Thus, considering the age range (65–98 years old) of the sample, age, as well as education and personality may predict SM and OM.

This study investigated how multifaceted depressive symptoms and personality traits contributed to associations between SM and OM. Haavisto and Boron (2018) deemed the lack of associations between SM and OM to individuals’ over-or underestimation of their memory abilities. Previous studies (Barnhoper & Chittka, 2010; Merema et al., 2013) indicated that personality traits and depressive symptoms may serve as confounding factors to one another. For example, some dysphoric mood may overlap with neurotic traits, whereas extroverted people may feel more energetic than introverted people. One of the strengths of this study was that the tested model was constructed to include the potential overlapping impact of personality traits on SM and OM in order to investigate the associations of depressive symptoms with SM and OM. Thus, the current results could be critical in practice for clinicians to determine and distinguish symptoms of pseudodementia or early stages of mild cognitive impairment in older adults.

Although the GDS factors provide the most predictive power for SM and OM, individual difference factors, such as personality, can further contribute to understanding these differences. Future research should take other individual differences (e.g., stress, coping skills, health behaviors, and social activity) into account to foster an accurate prediction of SM and OM performance, although it is clear that depressive symptoms, as measured by these GDS factors, should be included in future models. Moreover, testing effects of depressive symptoms and personality traits on SM and OM in longitudinal studies may be ideal. Depressive symptoms as time-varying factors may be experienced as a state, or temporary feeling, but may or may not resolve over time. In contrast, personality traits have been demonstrated to be stable across adulthood once individuals reach approximately 30 years of age (Costa & McCrae, 1988). Thus, future research in longitudinal studies distinguishing between time-varying and time-invariant variables can be made to examine the relationships between depression, personality, SM, and OM.

Footnotes

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Wonjeong Haavisto, PhD, is an adjunct professor in the Department of Gerontology at the University of Nebraska at Omaha. Her research focuses on the differential impact of psychological factors on subjective and objective memory in adulthood by using quantitative analyses.

Julie Blaskewicz Boron, PhD, is an associate professor and Doctoral Program Chair in the Department of Gerontology, and an affiliate associate professor in the Department of Biomechanics at the University of Nebraska at Omaha. Her research focuses on cognitive change in adulthood, and person-level indicators of change, such as personality, cognitive training, gait, and health behaviors by analyzing both primary and secondary data.

References

Allen

P. A.

Kaut

Baena

Lien

M. C.

Ruthruff

(2011). Individual differences in positive affect moderate age-related decline in episodic long-term memory. Journal of Cognitive Psychology, 23, 768–779. doi:10.1080/20445911.2011.570254

Asparouhov

Muthén

(2010). Weighted least squares estimation with missing data. Mplus Technical Appendix, 2010, 1–10.

Ayotte

B. J.

Potter

G. G.

Williams

H. T.

Steffens

D. C.

Bosworth

H. B.

(2009). The moderating role of personality factors in the relationship between depression and neuropsychological functioning among older adults. International Journal of Geriatric Psychiatry: A Journal of the Psychiatry of Late Life and Allied Sciences, 24(9), 1010–1019. doi:10.1002/gps.2213

Baker

T. J.

Bichsel

(2006). Personality of intelligence: Differences between young and cognitively healthy older adults. Personality and Individual Differences, 41(5), 861–871. doi:10.1016/j.paid.2006.02.017

Barnhoper

Chittka

(2010). Cognitive reactivity mediates the relationship between neuroticism and depression. Behaviour Research and Therapy, 48, 275–281. doi:10.1016/j.brat.2009.12.005

Bentz

B. G.

Hall

J. R.

(2008). Assessment of depression in a geriatric inpatient cohort: A comparison of the BDI and GDS. International Journal of Clinical and Health Psychology, 8(1), 93–104.

Boyle

L. L.

Lyness

J. M.

Duberstein

P. R.

Karuza

King

D. A.

Messing

(2010). Trait neuroticism, depression, and cognitive function in older primary care patients. The American Journal of Geriatric Psychiatry, 18(4), 305–312.

Brown

T. A.

(2014). Confirmatory Factor Analysis for Applied Research (2nd ed.). New York, NY: The Guilford Press.

Chapman

B. P.

Duberstein

P. R.

Sörensen

Lyness

J. M.

(2007). Gender differences in Five Factor Model personality traits in an elderly cohort. Personality and Individual Differences, 43(6), 1594–1603. doi:10.1016/j.paid.2007.04.028

10.

Costa

P. T.

McCrae

R. R.

(1988). Personality in adulthood: A six-year longitudinal study of self-reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology, 54(5), 853.

11.

Costa

P. T.

McCrae

R. R.

(1992a). Normal personality assessment in clinical practice: The NEO personality inventory. Psychological Assessment, 4(1), 5–13.

12.

Costa

P. T.

McCrae

R. R.

(1992b). Manual for the revised NEO personality inventory (NEO-PI-R) and NEO five-factor inventory (NEO-FFI). Odessa, FL: Psychological Assessment Resources.

13.

Curtis

R. G.

Windsor

T. D.

Soubelet

(2015). The relationship between Big-5 personality traits and cognitive ability in older adults – A review. Aging, Neuropsychology, and Cognition, 22(1), 42–71. doi:10.1080/13825585.2014.888392

14.

Cuttler

Graf

(2007). Personality predicts prospective memory task performance: An adult lifespan study. Scandinavian Journal of Psychology, 48, 215–231. doi:10.1111/j.1467=9450.2007.00570.x

15.

Folstein

M. F.

Folstein

S. E.

McHugh

P. R.

(1975). Mini-mental state”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189–198. doi:10.1016/0022-3956(75)90026-6

16.

Gerstorf

Herlitz

Smith

(2006). Stability of sex differences in cognition in advanced old age: The role of education and attrition. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 61(4), 245–249. doi:10.1093/geronb/61.4.P245

17.

Gilewski

M. J.

Zelinski

E. M.

(1988). Memory Function Questionnaire (MFQ). Psychopharmacology Bulletin, 24(4), 665–670.

18.

Gilewski

M. J.

Zelinski

E. M.

Schaie

K. W.

(1990). The Memory Functioning Questionnaire for assessment of memory complaints in adulthood and old age. Psychology and Aging, 5, 482–490. doi:10.1037/0882-7974.5.4.482

19.

Graham

E. K.

Lachman

M. E.

(2012). Personality stability is associated with better cognitive performance in adulthood: Are the stable more able? The Journals of Gerontology, 67B, 545–554. doi:10.1093/geronb/gbr149

20.

Gregory

Nettelbeck

Wilson

(2010). Openness to experience, intelligence, and successful aging. Personality and Individual Differences, 48, (2010). 895–899. doi:10.1016/j.paid.2010.02.017

21.

Haavisto

Boron

J. B.

(2018). Examination of discrepancies between subjective and objective memory with latent factors of the Geriatric Depression Scale. Aging and Mental Health. Advance online publication. doi:10.1080/13607863.2018.1428939

22.

Havins

W. N.

Massman

P. J.

Doody

(2012). Factor structure of the Geriatric depression scale and relationships with cognition and function in Alzheimer’s disease. Dementia and Geriatric Cognitive Disorders, 34, 360–372. doi:al.1159/000345787

23.

Hülür

Hertzog

Pearman

Ram

Gerstorf

(2014). Longitudinal associations of subjective memory with memory performance and depressive symptoms: Between-person and within-person perspectives. Psychology and Aging, 29(4), 814. doi:10.1037/a0037619

24.

Jessen

Wiese

Cvetanovska

Fuchs

Kaduszkiewicz

Koelsch

Werle

(2007). Patterns of subjective memory impairment in the elderly: Association with memory performance. Psychological Medicine, 37(12), 1753–1762.

25.

Jonker

Gerrlings

M. I.

Schmand

(2000). Are memory complaints predictive for dementia? A review of clinical and population based studies. International Journal of Geriatric Psychiatry, 15, 983–991. doi:10.1002/1099-1166(200011)15;11<983::AID-GPS238>3.0.CO;2-5

26.

Jonker

Launer

L. J.

Hooijer

Lindeboom

(1996). Memory complaints and memory impairment in older individuals. Journal of the American Geriatrics Society, 44(1), 44–49. doi: 10.1111/j.1532-5415.1996.tb05636.x

27.

Jungwirth

Fischer

Weissgram

Kirchmeyr

Bauer

Tragl

K. H.

(2004). Subjective memory complaints and objective memory impairment in the Vienna‐Transdanube aging community. Journal of the American Geriatrics Society, 52(2), 263–268. doi:10.111/j.10.1111/j.1532-5415.2004.52066.x

28.

Kahn

R. L.

Zarit

S. H.

Hilbert

N. M.

Niederehe

(1975). Memory complaint and impairment in the aged: The effect of depression and altered brain function. Archives of General Psychiatry, 32(12), 1569–1573. doi:10.1001/archpsyc.1975.01760300107009

29.

Lam

L. C.

Lui

V. W.

Tam

C. W.

Chiu

H. F.

(2005). Subjective memory complaints in Chinese subjects with mild cognitive impairment and early Alzheimer’s disease. International Journal of Geriatric Psychiatry, 20(9), 876–882. doi:10.1002/gps.1370

30.

Lane

C. J.

Zelinski

E. M.

(2003). Longitudinal linear models of the Memory Functioning Questionnaire. Psychology and Aging, 18(1), 38–53. doi: 10.1037/0882-7974.18.1.38

31.

Lewis

K. L.

Zelinski

E. M.

(2010). List and text recall differ in their predictors: Replication over samples and time. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 65(4), 449–458. doi:10.1093/geronb/gbq034

32.

Meier

Perrig-Chiello

Perrig

(2002). Personality and memory in old age. Aging, Neuropsychology, and Cognition, 9(2), 135–144.

33.

Merema

M. R.

Speelman

C. P.

Foster

J. K.

Kaczmarek

E. A.

(2013). Neuroticism (not depressive symptoms) predicts memory complaints in some community-dwelling older adults. The American Journal of Geriatric Psychiatry, 21(8), 729–736.

34.

Minett

T. S. C.

Da Silva

R. V.

Ortiz

K. Z.

Bertolucci

P. H. F.

(2008). Subjective memory complaints in an elderly sample: A cross‐sectional study. International Journal of Geriatric Psychiatry: A Journal of the Psychiatry of Late Life and Allied Sciences, 23(1), 49–54. doi: 10.1002/gps.1836

35.

Muthén

L. K.

Muthén

B. O.

(1998–2015). Mplus User’s Guide (7th ed.). Los Angeles, CA: Muthén & Muthén.

36.

Neupert

S. D.

Mroczek

D. K.

Spiro

III. (2008). Neuroticism moderates the daily relation between stressors and memory failures. Psychology and Aging, 23(2), 287–296. doi:10.1037/0822-7974.23.2.287

37.

O'Connor

D. W.

Pollitt

P. A.

Roth

Brook

C. P. B.

Reiss

B. B.

(1990). Memory complaints and impairment in normal, depressed, and demented elderly persons identified in a community survey. Archives of General Psychiatry, 47(3), 224–227. doi:10.1001/archpsyc.1990.01810150024005

38.

Okereke

O. I.

Kang

J. H.

Cook

N. R.

Gaziano

J. M.

Manson

J. E.

Buring

J. E.

Grodstein

(2008). Type 2 diabetes mellitus and cognitive decline in two large cohorts of community‐dwelling older adults. Journal of the American Geriatrics Society, 56(6), 1028–1036. doi:10.1111/j.1532-5415.2008.01686.x

39.

Pearman

(2009). Basic cognition in adulthood: Combined effects of sex and personality. Personality and Individual Differences, 47, 357–362. doi:10.1016/j.paid.2009.04.003

40.

Pearman

Hertzog

Gerstorf

(2014). Little evidence for links between memory complaints and memory performance in very old age: Longitudinal analyses from the Berlin Aging Study. Psychology and Aging, 29(4), 828–842. doi:10.1037/a0037141

41.

Pearman

Storandt

(2004). Predictors of subjective memory in older adults. Journal of Gerontology: Psychological Sciences, 59(1), P4–P6.

42.

Pearman

Storandt

(2005). Self-discipline and self-consciousness predict subjective memory in older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 60(3), P153–P157.

43.

Pocnet

Rossier

Antonietti

J. P.

von Gunten

(2013). Personality features and cognitive level in patients at an early stage of Alzheimer’s disease. Personality and Individual Differences, 54(2), 174–179. doi:10.1016/j.paid.2012.08.035

44.

Saxby

B. K.

Harrington

McKeith

I. G.

Wesnes

Ford

G. A.

(2003). Health Psychology, 22(6), 587–591. doi:10.1037/0278-6133.22.6.587

45.

Schaie

K. S.

Willis

S. L.

Caskie

G. I. L.

(2004). The Seattle Longitudinal Study: Relationship between personality and cognition. Aging Neuropsychology and Cognition, 11(2-3), 304–324. doi:10.1080/13825580490511134

46.

Slavin

M. J.

Brodaty

Kochan

N. A.

Crawford

J. D.

Trollor

J. N.

Draper

Sachdev

P. S.

(2010). Prevalence and predictors of “subjective cognitive complaints” in the Sydney Memory and Ageing Study. The American Journal of Geriatric Psychiatry, 18(8), 701–710.

47.

Soubelet

Salthouse

T. A.

(2011). Personality-cognition relations across adulthood. Developmental Psychology, 47, 303–310. doi:10.1037/a0021816

48.

Steffens

D. D.

Potter

G. G.

(2008). Geriatric depression and cognitive impairment. Psychological Medicine, 38, 163–175. doi:10.1017/S003329170700102X

49.

Terry

P. T.

Puente

A. N.

Brown

C. L.

Faraco

C. C.

Miller

L. S.

(2013). Openness to experience is related to better memory ability in older adults with questionable dementia. Journal of Clinical and Experimental Neuropsychology, 35(5), 509–517. doi:10.1080/13803395.2013.795932

50.

Weber

Giannakopoulos

Bacchetta

J. P.

Quast

Herrmann

F. R.

Delaloye

Canuto

(2012). Personality traits are associated with acute major depression across the age spectrum. Aging & Mental Health, 16(4), 472–480. doi.org/10.1080/13607863.2011.630375

51.

Wilson

R. S.

Evans

D. A.

Bienias

J. L.

De Leon

C. M.

Schneider

J. A.

Bennett

D. A.

(2003). Proneness to psychological distress is associated with risk of Alzheimer’s disease. Neurology, 61(11), 1479–1485. doi:10.1212/01.WNL.0000096167.56734.59

52.

Wilson

R. S.

Schneider

J. A.

Boyle

P. A.

Arnold

S. E.

Tang

Bennett

D. A.

(2007). Chronic distress and incidence of mild cognitive impairment. Neurology, 68(24), 2085–2092. doi:10.1212/01/wml.0000264930.97061.82

53.

Yesavage

J. A.

Brink

T. L.

Rose

Lum

Huang

Adey

Leirer

(1983). Development and validation of a geriatric depression screening scale: A preliminary report. Journal of Psychiatric Research, 17(1), 37–49.

54.

Zelinski

E. M.

Burnight

K. P.

Lane

C. J.

(2001). The relationship between subjective and objective memory in the oldest old: Comparisons of findings from a representative and a convenience sample. Journal of Aging and Health, 13(2), 248–266. doi:10.1177/089826430101300205

55.

Zelinski

E. M.

Gilewski

M. J.

(2004). A 10-item Rasch modeled memory self-efficacy scale. Aging & Mental Health, 8(4), 293–306. doi:10.1080/13607860410001709665

56.

Zelinski

E. M.

Gilewski

M. J.

Anthony-Bergstone

C. R.

(1990). Memory functioning questionnaire: Concurrent validity with memory performance and self-reported memory failures. Psychology and Aging, 5, 388–399.

57.

Zelinski

E. M.

Kennison

R. F.

(2011, June). Long Beach Longitudinal Study. ICPSR26561-v2. Ann Arbor, MI: Inter-University Consortium for Political and Social Research [ Distributor]. doi: 10.3886/ICPSR26561.v2

58.

Zelinski

E. M.

Kennison

R. F.

Watts

Lewis

K. L.

(2009). Convergence between cross-sectional and longitudinal studies: Cohort matters. In Bosworth

Hertzog

(Eds). Aging and cognition: Research methodologies and empirical advances (pp. 101–118). Washington, DC: American Psychological Association.

59.

Zelinski

E. M.

Lewis

K. L.

(2003). Adult age differences in multiple cognitive functions: Differentiation, dedifferentiation, or process-specific change? Psychology and Aging, 18(4), 727–745. doi:10.1037/0882-7974.18.4.727

60.

Zimprich

Martin

Kliegel

(2003). Subjective cognitive complaints, memory performance, and depressive affect in old age: A change-oriented approach. The International Journal of Aging and Human Development, 57(4), 339–366. doi:10.2190/G0ER-ARNM-BQVU-YKJN

61.

Zunzunegui

M. V.

Minicuci

Blumstein

Noale

Deeg

Jylhä

Clesa Working Group, (2007). Gender differences in depressive symptoms among older adults: A cross-national comparison. Social Psychiatry and Psychiatric Epidemiology, 42(3), 198–207. doi:10.1007/s00127-007-0158-3