Age differences in everyday problem solving

Abstract

Reductions in everyday problem solving (EPS) are often reported in older age, although it has been suggested that problem context may modify this effect. We evaluated the impact of two aspects of problem context: age appropriateness (age-neutral vs. older-age content) and problem type (interpersonal vs. practical) on EPS performance in 175 adults aged 18–87. Older adults generated fewer solutions to ill-structured EPS vignettes than younger and middle-aged adults. Middle-aged adults demonstrated an advantage on practical problems. While all age groups demonstrated a relative performance advantage for interpersonal content on older age problems, older adults showed the least relative benefit in this condition. Thus older adults do not exhibit relative performance gains on EPS problems designed to be most salient and relevant to this age group.

Keywords

applied cognition cognition executive function life course and developmental change problem solving

Cognitive declines are often reported on traditional laboratory measures of problem solving and memory in later life; however, questions remain regarding to what extent and under what conditions everyday problem solving (EPS) may be robust to the effects of age. In contrast to traditional approaches for assessing cognition, EPS tasks are designed to assess the ability to solve specific problematic situations that are representative of what one might face in daily life (Denney, 1990). While there are several approaches to the study of EPS in adulthood, attempts to consolidate this literature (Marsiske & Margrett, 2006; Thornton & Dumke, 2005) have provided robust evidence for age differences favouring younger adults on EPS performance measures, which emphasize the fluency, accuracy, and/or effectiveness of participants’ responses. These EPS performance measures include well-structured problems (for which the initial state and desired end-state are provided; e.g., Allaire & Marsiske, 1999), ill-structured vignettes (requiring the generation of multiple safe and effective solutions to open-ended problems; e.g. Denney & Pearce, 1989), and those emphasizing the accuracy/effectiveness of the generated solution/s (e.g., Crawford & Channon, 2002). In contrast, methods that allow participants to rate their own perceived problem-solving approach (e.g., Cornelius & Caspi, 1987), or emphasize qualitative aspects of the problem approach (e.g., Berg, Strough, Calderone, Sansone, & Weir, 1998), often yield different pictures of EPS in later life (for reviews, see Marsiske & Margrett, 2006; Thornton & Dumke, 2005).

In the current study, we examine the influence of contextual factors on EPS performance using an ill-structured vignette task, on which age differences favouring younger adults are typically reported (see Marsiske & Margrett, 2006; Thornton & Dumke, 2005). These measures approximate “real-world” problem solving by requiring the participant to generate spontaneous, safe and effective solutions to common problems as they emerge. Importantly, there is increasing evidence linking performance on these EPS measures to real-world outcomes. For example, performance using ill-structured EPS measures predicts quality of life in older adults beyond that accounted for by traditional cognitive tasks (Gilhooly et al., 2007). Furthermore, performance on these measures is a unique predictor of life skills functioning (Thornton, Kristinsson, DeFreitas, & Thornton, 2010), and a better predictor of medication adherence (Gelb, Shapiro & Thornton, 2010) than traditional cognitive measures. Similarly, in older adults, performance on well-structured EPS performance measures explains unique variance in self-rated functioning (Allaire & Marsiske, 2002), and is predictive of increased risk of mortality (Allaire & Willis, 2006; Weatherbee & Allaire, 2008). Given the importance of effective problem solving in maintaining independence and quality of life, it is critical to determine the factors associated with optimal EPS performance.

EPS performance in context

It has been suggested that experience and accumulated knowledge in later life may bolster everyday cognition when the to-be-solved problem is familiar and relevant to the older adult (Baltes, 1993) and that increasing value is placed on emotionally meaningful relationships in later life (Carstensen & Mikels, 2005). Within this framework, one may predict that older adults have more experience with and greater affinity for interpersonal problems, and this may translate into context-specific EPS performance gains. Findings from studies examining age differences in strategy selection appear to support this contention (e.g., Blanchard-Fields, 2007; Blanchard-Fields, Mienaltowski, & Seay, 2007); however, findings from studies using performance-based measures are less clear. For example, previous studies have compared age differences in solution generation on interpersonal (e.g., social predicaments) and practical EPS problems (e.g., household repairs), and have reported worse performance in older adults in both domains (Heidrich & Denney, 1994; Strough, McFall, Flinn, & Schuller, 2008). Nonetheless, a meta-analysis of this literature found that age differences on EPS performance measures were attenuated (but not eliminated) when the problem type reflected interpersonal as compared to practical concerns (Thornton & Dumke, 2005). Thus while absolute EPS performance may not be equivalent across age regardless of problem domain, the question remains whether older adults demonstrate relative performance gains for interpersonal problems on performance-based measures.

The age-appropriateness of the to-be-solved problem has also been suggested as a potential modifier of age differences on EPS performance measures. Researchers have postulated that performance gains should be evident in later life when the problem context reflects age-appropriate concerns (e.g., retirement, widowhood), although findings are disparate. While some authors report worse EPS performance (as indexed by solution generation) in later life despite using older-age problems (Denney & Pearce, 1989; Denney, Pearce, & Palmer, 1982), others report an advantage for older adults using age-appropriate problems (Artistico, Cervone, & Pezzuti, 2003). One explanation for these disparities may be that previous studies have typically confounded the age-appropriate and interpersonal domains in their problem sets, precluding the ability to examine each domain separately. Thus while there is evidence suggesting that problem domain may prove an important modifier of age differences on EPS performance measures, we are aware of no study to date that has systematically manipulated the impact of both age-appropriateness and problem type.

Another important question involves how EPS performance in mid-life may be affected by problem context. Several studies have reported that EPS performance (as indexed by solution generation) peaks in middle age, followed by a decline in later years (Denny & Palmer, 1981; Denney & Pearce, 1989; Denney et al., 1982), although this trend is not universally reported (see Thornton & Dumke, 2005). It is often presumed that EPS is compiled from basic mental abilities that tend to decline in later life (Marsiske & Willis, 1995), and this view is supported by associations between performances on everyday and traditional psychometric measures of memory, working memory, and executive functioning (e.g., Allaire & Marsiske, 1999; Kirasic, Allen, Dobson, & Binder, 1996; Thornton, Deria, Gelb, Hill, & Shapiro, 2007). Thus the increase in EPS performance reported in some middle-aged samples has been attributed to the additive effects of experience in solving everyday problems and the relative preservation of cognitive capacities (Denney, 1990). We have previously shown that performance on traditional neuropsychological measures (i.e., memory/executive functioning and vocabulary/verbal fluency) partially (but not fully) mediates the associations between age and EPS (Thornton et al., 2007). Therefore we assessed whether age differences across problem context were maintained after statistically controlling for performance on traditional cognitive measures.

Current aims

Using ill-structured vignettes derived from the extant literature, we evaluated the impact of two aspects of problem context, age-appropriateness (age-neutral vs. older-age content), and problem type (interpersonal vs. practical concerns) on EPS performance in younger, middle-aged, and older adults. We predicted that, while older adults would generate fewer EPS solutions overall than either younger or middle-aged adults, they would show enhanced performance gains (i.e., a relatively greater number of safe and effective solutions) on the very problems presumed to be most salient to older adults (older-age interpersonal problems), and greater relative performance costs (i.e., generate fewer solutions) on the problems presumed least salient (age-neutral practical problems). Toward these ends, we first developed our protocol by having younger and older adults rate EPS vignettes taken from the extant literature along these dimensions. We then addressed whether adult age differences in EPS performance were impacted by these contextual factors in a new sample of 175 participants.

Method

Participants

We recruited middle-aged (ages 51–64) and older adults (ages 65–87) through advertisements at community centres throughout the greater Vancouver, British Columbia (BC) area. Younger participants (ages 18–30) were recruited though the Simon Fraser University (SFU) undergraduate participant pool. To ensure that participants could adequately comprehend the testing protocol, exclusion criteria included a history of major illnesses with known direct central nervous system (CNS) effects, previously identified cognitive impairments (e.g., diagnosis of dementia), a current diagnosis of a major psychiatric illness, or visual acuity less than 20/50. The protocol was approved by the SFU research ethics board, and all participants provided written consent. Information regarding demographic characteristics and mean performances on the variables of interest is presented in Table 1.

Table 1.

Demographic and EPS variables

Participant characteristics	Younger (n = 84)	Middle-aged (n = 46)	Older (n = 45)	p-value*
Age (mean ± SD)	^a19.52 ± 2.18	^b59.35 ± 3.45	^c71.69 ± 5.68	<.001
Female (%)	59 (70%)	35 (76%)	32 (71%)	n.s.
Education (mean yrs. ± SD)	^b13.13 ± 1.26	^a15.52 ± 2.35	^b13.93 ± 2.64	<.001
EPS total score	^a63.49 ± 14.26	^b72.65 ± 20.82	^c54.44 ± 15.45	<.001
EPS practical overall	^b3.43 ± .80	^a4.28 ± 1.22	^b3.20 ± .85	<.001
EPS interpersonal overall	^b4.50 ± 1.17	^b4.80 ± 1.49	^a3.60 ± 1.18	<.001
EPS age neutral overall	^a3.99 ± .96	^b4.57 ± 1.24	^c3.42 ± 1.01	<.001
EPS older age overall	^a3.95 ± .95	^b4.51 ± 1.45	^c3.38 ± 1.13	<.001

Notes. EPS = everyday problem solving; * = p-value obtained from ANOVA, planned comparison follow-up tests, or Pearson chi-square; ^{a, b, c} = level at which group differences occur (e.g., ^{a b c} = all groups different from each other; ^{a b b} = ^a different from ^b, which are not different from each other).

Materials

EPS task

A preliminary aim was to identify a set of problems that could represent the problem-solving domains of interest. To develop our protocol, we contacted 20 older adults who had participated in previous studies in our lab (mean age = 73.05, SD = 8.75), and asked them to answer a series of questions based upon a set of ill-structured EPS vignettes that have been used extensively in adult developmental studies of EPS (Artistico et al., 2003; Denney & Palmer, 1981; Denney & Pearce, 1989; Haught, Hill, Nardi, & Walls, 2000; Haught & Walls, 2007; Marsiske & Willis, 1995). Twenty younger participants were recruited using the university undergraduate participant pool (mean age = 19.35, SD = 1.18). Years of education did not differ significantly between younger (M = 14.35, SD = 1.18) and older adults (M = 14.60, SD = 2.10), and the gender distributions were equivalent. Older participants were contacted via letter and asked to rate 20 problem vignettes along five dimensions: (a) “In your opinion, how important would this problem be in a person’ s life?”; (b) “How important would this problem be to you?”; (c) “How often have you had to deal with a similar problem in your own life?”; (d) “In your opinion, what age group would more commonly face this problem?”; and (e) “In your opinion, does this problem reflect more of a practical (e.g., household/mechanical oriented) or social (e.g., relationship/caregiver oriented) concern?” Participants responded on a five-point scale ranging from 1 (not at all) to 5 (very much) for questions (a) through (c). Responses for question (d) ranged from 1 (younger adults only) to 5 (older adults only). Responses for question (e) ranged from 1 (completely practical) to 5 (completely social). Younger adults were tested either individually or in small groups in the Cognitive Aging Laboratory at SFU.

We then examined mean ratings to determine a subset of 16 from the original 20 problems with four from each category of interest. For question (d), it was clear that few problems were rated as primarily “younger adult” problems. Rather, responses differentiated best between age-neutral problems (“occurs equally at any age”) and older-age problems. Ratings were compared using independent and paired samples t-tests. Older-age problems were rated as more likely to be faced by older adults (M = 3.90, SD = .24) than were age-neutral problems (M = 3.14, SD = .25; t(39) = 15.90, p < .001), and the magnitude of this difference was large (d = 3.10; Cohen, 1992). Furthermore, problems identified as having a practical focus (M = 2.25, SD = .42) were rated differently than those identified as interpersonal (M = 4.05, SD = .48; t(39) = 18.34, p < .001) and the magnitude of this difference was also large (d = 4.00; Cohen, 1992). These differences were maintained in both younger and older participants (all ps < .001). Older participants rated all problem domains as more important both generally and personally than did younger participants (all ps < .05). Older and younger participants did not differ in their ratings of personal experience with problem domains. Mean ratings for each problem type are available upon request. The EPS task used for the current study consisted of the 16 paper-and-pencil vignettes drawn from the preliminary study. We used four problems which showed the highest discrimination from within each of four categories: (a) age-neutral practical; (b) age-neutral interpersonal; (c) older-age practical; and (d) older-age interpersonal. The final problem set for each condition is presented in the Appendix.

Measures of neuropsychological functioning

Verbal memory was measured with the California Verbal Learning Test—2 (CVLT-II; Delis, Kramer, Kaplan, & Ober, 2000), which assesses learning over repeated trials, susceptibility to memory interference, and delayed verbal memory. Executive functioning measures were the trail making, colour-word interference and category fluency subtests of the Delis-Kaplan Executive Functioning System (D-KEFS; Delis, Kaplan, & Kramer, 2001), which examine three aspects of executive functioning: mental set shifting, cognitive inhibition, and response monitoring (Miyake et al., 2000). Vocabulary was assessed with the Vocabulary (V-2) and Extended Range Vocabulary (V-3) subtests from the ETS kit (Ekstrom, Harmon, & Derman, 1976).

Procedures

All participants were tested individually in a two-hour session at local libraries, community centres, or the SFU Cognitive Aging Laboratory. Middle-aged and older participants received monetary compensation for their time and travel, and younger participants received course credit that was predetermined and not influenced by actual time on task. The measures were administered and scored by trained research assistants according to standardized procedures. One problem was presented per page, and problems were randomly ordered. Two versions of the problem sets were created (version 2 presented the second half of the problem set first), and test version was randomly assigned to participants. None of the participants involved in rating the initial items in the preliminary study were involved in the main study of 175 adults. Participants were asked to read each problem carefully and to write down as many solutions as possible, even if it was a solution that they themselves would not adopt. Participants were given no time limits for completion of this task, and few took longer than the time allotted for the full assessment (two hours).

The scoring criteria for the EPS task were devised (Denney & Pearce, 1989) and adapted by previous authors (Marsiske & Willis, 1995) to incorporate both an individual’s wealth and quality of ideas. To receive a point, a solution had to satisfy the following criteria: (1) dealt directly with the problem at hand; (2) safe for all individuals involved in the problem; and (3) effective in resolving the problem for both the short and long term. The total number of conceptually distinct safe and effective solutions generated by each participant for each problem was combined into a total EPS score. In our laboratory, inter-rater agreement using these criteria was determined to be very high (r_ic = .96). The percentage of safe and effective solutions as compared to total number of solutions generated (pure fluency) was comparable between middle-aged (M = 85%, SD = .05) and older participants (M = 88%, SD = .05; p nonsignificant [n.s.]), whereas the percentage was lower for younger participants than either of the older age groups (M = 82%, SD = .06); F(2, 172) = 16.74, p < .001.

Statistical analyses

Analyses were conducted using SPSS 18 software (SPSS Inc. Chicago, IL). Group differences across demographic and cognitive variables were examined with analysis of variances (ANOVAs) or nonparametric tests where appropriate. Since detailing neuropsychological functioning was not a focus of the current study, we conducted a principal component analysis with the neuropsychological test scores to reduce the number of variables. This revealed two components with eigenvalues exceeding 1.0, explaining 47% and 19% of the variance, respectively. On the first component, labelled “memory and executive functions,” measures of learning and memory, mental set shifting, and cognitive inhibition loaded highly. The second component consisted of the vocabulary and category fluency measures, and is considered the “vocabulary/fluency” factor. These two scores were retained for use as potential covariates in subsequent analyses.

To determine which variables would be included as covariates in the primary analysis, we first examined the associations between the variables of interest (see Table 2). To reduce the possibilities of capitalizing on chance associations, we chose to add only variables to the model that resulted in group differences at a level of p < .01. As seen in Table 1, middle-aged adults had a higher level of education than either younger or older adults. Higher education and better neuropsychological functioning (memory/executive and vocabulary/fluency abilities) were significantly associated with better overall EPS performance (see Table 2) and were included as covariates in separate analyses. With these covariates in the model, the relationships of interest were only mildly attenuated, and the main effects and interactions reported herein all retained significance (i.e., p < .05). Thus we dropped the covariates from the final model so that we could present the unadjusted means and simplify the interpretation of the results.

Table 2.

Intercorrelations among cognitive and demographic variables of interest (n = 175)

	Variable	1	2	3	4	5	6	7	8
1	Age	—
2	Education	.35⁺	—
3	EPS Total	−.06	.25⁺	—
4	EPS: NP	.14	.26⁺	.84⁺	—
5	EPS: OP	−.04	.24⁺	.81⁺	.64⁺	—
6	EPS: NI	−.26⁺	.16*	.85⁺	.60⁺	.56⁺	—
7	EPS: OI	−.09	.21⁺	.91⁺	.63⁺	.67⁺	.71⁺	—
8	Mem/Exec	−.62⁺	−.21⁺	.28⁺	.13	.21⁺	− .38⁺	.23⁺	—
9	Vocabulary/Fluency	.46⁺	.37⁺	.37⁺	.45⁺	.33⁺	.20⁺	− .31⁺	.44⁺

Notes. Correlation is significant at *p < .05, ⁺ p < .01 (two-tailed); EPS = everyday problem solving; EPS NP = age-neutral practical problems; EPS OP = older-age practical problems; EPS NI = age-neutral interpersonal problems; EPS OI = older-age interpersonal problems; mem/exec = memory/executive functioning factor; vocabulary/fluency = crystallized abilities factor.

To address our primary prediction that older adults would show enhanced performance gains on problems that have been suggested to be most salient to older adults (older-age interpersonal problems), and greater performance costs on problems presumed least salient (age-neutral practical problems), we conducted repeated measures ANOVA, with EPS performance (mean number of safe and effective solutions in each of the four conditions) as the dependent variable. We utilized a 3 (age group) × 2 (problem type) × 2 (age-appropriateness) mixed linear model in which age group was the between-subject factor, and problem type (interpersonal vs. practical) and age-appropriateness (age-neutral vs. older-age problems) were within-subject factors. Mean contrasts were performed to decompose the significant main effects and interactions as appropriate.

Results

As predicted, older adults generated fewer solutions (M = 3.40, SE = .15) to the EPS problems overall, F(2,172) = 13.85, p < .001, than either middle-aged (M = 4.54, SE = .15) or younger adults (M = 3.97, SE = .11). In addition, the number of responses was higher for interpersonal (M = 4.30, SE = .10) than practical problems (M = 3.64, SE = .07) for all age groups, F(1, 172) = 97.03, p < .001. However, these relationships were impacted by problem type. Examination of the interaction between problem type and age group, F(2, 172) = 11.21, p < .001, using planned comparisons revealed that middle-aged adults generated more solutions than both younger t(172) = 4.88, p < .001, and older adults, t(172) = −5.43, p < .001, on practical problems, whereas, counter to our predictions, older adults generated fewer solutions than both younger, t(172) = −3.85, p < .001, and middle-aged adults on interpersonal problems, t(172) = −4.52, p < .001. These means are presented in Table 1.

A significant age-appropriateness by problem type interaction was also revealed, F(1, 172) = 335.24, p < .001. Examination of this effect indicated that, for age neutral problems, all participants performed better on practical (M = 4.18, SE = .10) as compared to interpersonal problems (M = 3.80, SE = .10), whereas the opposite pattern emerged for older-age problems, on which participants performed better on interpersonal problems (M = 4.80, SE = .13) in comparison to practical problems (M = 3.10, SE = .07). These relationships were further qualified by a significant age-appropriateness by problem type by age-group interaction, F(2, 172) = 11.94, p < .001, which was decomposed by comparing the means for each group across each level of problem type and age-appropriateness using planned comparisons (see Figure 1). On practical problems, significant age-group differences emerged for both age-neutral, F(2, 172) = 18.67, p < .001, and older-age problems, F(2, 172) = 9.40, p < .001. For both conditions (age-neutral and older-age problems), middle-aged adults performed better than both older, t(172) = 5.33, p < .001; t(172) = 4.28, p < .001, respectively, and younger adults, t(172) = −5.46, p < .001; t(172) = −2.98, p < .005, respectively, while older and younger adults' performances were not significantly different. On interpersonal problems, age-group differences were again observed for both age-neutral, F(2, 172) = 12.14, p < .001, and older-age problems, F(2, 172) = 9.21, p < .001. However, for age-neutral interpersonal problems, older adults performed worse than both younger, t(172) = −4.64, p < .001, and middle-aged adults, t(172) = 4.02, p < .001, who exhibited equivalent performance. On older-age interpersonal problems, middle-aged adults performed better than both younger, t(172) = −2.11, p < .05, and older adults, t(172) = 4.28, p < .001, while older adults also performed worse than younger adults in this condition, t(172) = −2.76, p < .01.

Figure 1.

Mean EPS performance in each age group for each problem type.

To further decompose the age-appropriateness by problem type by age-group interaction, we computed the difference scores between interpersonal and practical problems and compared this score between age groups at both levels of age-appropriateness. By using these difference scores, we created a new dependent variable (relative benefit) that allowed us to examine relative performance gains for interpersonal problems within each age-appropriateness condition. A positive difference score indicated that participants performed relatively better on interpersonal versus practical problems, and a negative difference score was associated with better relative performance on practical problems. In contrast to our predictions, on older-age problems, older adults actually demonstrated less relative benefit for interpersonal problems than either younger, t(172) = − 2.11, p < .05, or middle-aged adults, t(172) = −2.25, p < .05), whereas younger and middle-aged adults showed equivalent interpersonal problem benefit. Furthermore, on age-neutral problems, younger adults demonstrated a relative benefit for interpersonal problems as compared with both middle-aged, t(172) = −6.47, p < .001, and older adults, t(172) = −4.64, p < .001, who showed the reverse pattern by generating more solutions to practical problems. The difference scores reflecting the comparison of relative benefits between the age-appropriateness conditions across age groups are presented in Figure 2.

Figure 2.

Relationships between age group, problem type, and age-appropriateness.

Discussion

The current findings suggest that problem domain does appear to influence EPS performance in adulthood, although the relationships are not straightforward. Consistent with our predictions, older adults generated fewer solutions to EPS problems overall than younger and middle-aged adults. However, counter to our predictions, their performance was not differentially bolstered under conditions presumed to be optimal for EPS performance in late life: when the problem domain involves older age content, and when the problem type reflects interpersonal concerns. While older adults did generate more solutions to interpersonal problems than to practical problems, this was true for all age groups, with younger adults actually showing the greatest relative benefit in this condition. On older-age problems, all participants generated more solutions when the problem content was interpersonal; however, older adults actually showed the least interpersonal problem benefit in this condition.

Consistent with some previous reports (e.g., Denney & Palmer, 1981; Denney & Pearce, 1989; Denney et al., 1982), we found that absolute EPS performance was greatest in middle-aged adults. In addition, a novel contribution of these findings is that EPS in mid-life was particularly bolstered on practical problems. Superior problem solving in mid-life has been previously attributed to the additive effects of experience in solving problems of daily life and relative preservation of cognitive capacities (Denney, 1990; Heidrich & Denney, 1994). In the current study this mid-life benefit was maintained even after statistically controlling for differences in education and performance on traditional neuropsychological tasks. Previous studies have shown that age differences in EPS are associated with, but not fully explained by, variations in basic mental abilities (Allaire & Marsiske, 1999; Thornton et al., 2007). Thus the important question remains: What factors underlie these robust age differences in EPS performance? Studies examining age differences with a variety of EPS measures and potential modifiers are needed to determine how robust the apparent mid-life boost in EPS performance may prove to be, and whether this boost can be extended into later life.

In terms of EPS performance in later life, the current findings offer little support for models suggesting that older adults may show relative performance gains on problems designed to be most salient and relevant to this age group. It has been suggested that interpersonal problem content may become more salient with increasing age, as evidenced by older adults’ strategy selection (Blanchard-Fields et al., 2007) and judgements of solution quality (Crawford & Channon, 2002). Nonetheless, the current findings are consistent with previous reports that this does not translate into performance gains as indexed by solution generation (Heidrich & Denney, 1994).

In addition, the current findings suggest that a straightforward explanation of age differences in terms of “age-appropriateness” is unlikely. These findings are in contrast with a previous report of a selective advantage in solution-generation for older adults on “older-adult” problems (Artistico et al., 2003). There is some indication that methodological or sample differences may underlie this disparity, as “interpersonal content” and “age-appropriateness”were confounded in this earlier study. Furthermore, our younger participants generated more than twice as many solutions to “older age” problems (avg. = 3.95 solutions) than younger participants in this previous study (avg. = 1.83 solutions; Artistico et al., 2003), despite the fact that similar instructions and scoring criteria were implemented (i.e., explicitly encouraging participants to generate as many solutions as possible, even if it is one that they may not adopt). In addition, the actor in the stimulus materials was explicitly identified as a “young adult/student” or an “older person” in the study by Artistico and colleagues. In the current study, the actor in each vignette was described only as “a person,” as we intended age relevance to be determined by the problem content alone. Perhaps as a result of this change, few problems were rated as primarily concerning a “younger adult” and were treated instead as “age neutral” problems. Future studies are needed to determine to what extent differences in problem wording may influence salience and/or EPS performance.

The current findings should be considered in light of certain limitations. It could be argued that our problem set, although rated by a separate group of older and younger adults for problem type and age-appropriateness, did not adequately capture the characteristics of “salience.” It is important to note, however, that the current problem set was comprised of ill-structured EPS vignettes that had been used in previous studies assessing age differences in either problem type or age-appropriateness (Artistico et al., 2003; Denney & Palmer, 1981; Denney & Pearce, 1989; Haught et al., 2000). Furthermore, older participants tended to rate all problem domains as more important both generally and personally than did younger participants. Notably, it appears that the older adult’s perception of problem importance did not translate into performance gains.

It is also important to note that there are several methods for assessing EPS, and controversy remains regarding how to best index performance in older adults. The ill-structured EPS measure used in the current study requires participants to come up with as many independent, practical, safe, and effective solutions to everyday problems as they can. This approach assumes that an individual who is able to generate numerous effective solutions is better able to flexibly apply their knowledge and experience to a problem. Some have argued that EPS methods that require solution-generation penalize older adults’ performance by emphasizing fluency, and may mischaracterize effective problem solving if older adults elect to use a more selective approach (Berg, Meegan, & Klaczynski, 1999). It is important to note that, in the current study, the percentage of safe and effective solutions versus total number of solutions was comparable in middle-aged and older adults. This argues against the notion that performance differences are due to older adults “holding back” and relying on a more selective approach. Regardless, it appears that whether or not individuals elect, or are less able, to generate effective strategies, the net result is that a higher number of effective strategies predicts better everyday functioning across a variety of domains (Gelb et al., 2010; Gilhooly et al., 2007; Thornton et al., 2010). Since the study of EPS is premised on the assumption of ecological validity (see Marsiske & Margrett, 2006), fluency-based EPS measures are likely well-suited for incorporation in studies examining real-world functioning in older populations. Future research is needed to determine what aspects of EPS performance are essential for optimal everyday functioning, and what factors underlie performance declines in later life.

In sum, by systematically examining the impact of both age-appropriateness and content (interpersonal vs. practical) on EPS domains, the current findings demonstrate that a simple explanation of age differences in terms of problem context is unlikely. The findings suggest that, while EPS performance in adulthood is influenced by both problem domain and problem type, there is little support for models suggesting that older adults show relative performance gains on problems suggested to be most salient and relevant to this age group. With mounting evidence for the utility of EPS performance measures in predicting real-world functioning, it is increasingly important to identify positive and negative modifiers of EPS in older age, and to determine how to maximize EPS effectiveness across the lifespan.

Footnotes

Acknowledgements

We thank Dr. Allen E. Thornton and Dr. Rachel Fouladi for their helpful comments on earlier versions of this manuscript. We thank Jessica Kubik for her assistance with data collection. We address special thanks to all those who agreed to participate in this study.

Funding

This work was supported in part by a Social Sciences and Humanities Research Council (SSHRC) Standard Research Grant awarded to the first author. Theone S. E. Paterson was supported by an SSHRC Canada Graduate Master’s Scholarship and by a Michael Smith Foundation for Health Research Training Award. Sophie E. Yeung was supported by an SSHRC Canada Graduate Master’s Scholarship.

Appendix.

Everyday problem-solving vignettes*

Problem type
Age-neutral practical
A person bought a vacuum cleaner from a door-to-door salesman. After two or three weeks, the vacuum cleaner no longer works. What should this person do? (Denney & Palmer, 1981; Haught et al., 2000; Haught & Walls, 2007).
One evening, a person goes to the refrigerator and notices that it is not cold inside but, rather, it’s warm. What should the person do? (Denney & Palmer, 1981; Haught & Walls, 2007).
A person lives in a house with a basement. One night there is a flash flood and they notice that the basement is being flooded by the water coming in the window wells. What should they do? (Denney & Palmer, 1981; Haught & Walls, 2007).
If someone was travelling by car and got stranded out on a highway during a blizzard, what should they do? (Denney & Palmer, 1981; Haught et al., 2000; Haught & Walls, 2007).
Age-neutral interpersonal
A woman is taking care of an 8-year-old child. One day she arrives at home 15 minutes after the child typically comes home on the school bus, but the child is not there. After an hour and 30 minutes, the woman has still not heard from her. It’s beginning to get dark. What should she do? (Denney & Palmer, 1981; Haught et al., 2000; Haught & Walls, 2007).
A person who avoids social situations because of extreme shyness wants to change this. What should he/she do? (Artistico et al., 2003).
A person is taking care of a friend’s or relative’s child while the parents are travelling in Europe. The child gets hurt and needs medical attention, but when the person takes the child to the emergency room, they find that the hospital refuses to do anything for the child until they have a parent’s signature. What should the person do? (Denney & Palmer, 1981; Haught & Walls, 2007).
A man is experiencing a very difficult time with his partner, and wants to do something to improve the relationship. What should he do? (Artistico et al., 2003).
Older-age practical
Let’s say that a man’s doctor has told him to take it easy because of a heart condition. It’s the summertime and the man’s yard needs to be mowed but the man cannot afford to pay someone to mow the lawn. What should he do? (Denney & Pearce, 1989; Marsiske & Willis, 1995).
A woman can drive her car to run errands except in winter when the weather is bad. What should she do about getting groceries and other necessities when the weather is bad? (Denney & Pearce, 1989; Marsiske & Willis, 1995).
A couple is living on a small pension and they have no other source of income. One winter they find that the heating bills are so high that they cannot pay them. What should they do? (Denney & Pearce, 1989; Marsiske & Willis, 1995).
Suppose that a woman needs to go somewhere at night. She cannot see well enough to drive at night and it’s too far to walk. What should she do? (Denney & Pearce, 1989; Marsiske & Willis, 1995).
Older-age interpersonal
A man has just retired. He doesn’t have any hobbies because he has never had time for them before. Now he is really bored. What should he do? (Denney & Pearce, 1989; Marsiske & Willis, 1995).
A woman feels that her sons are too intrusive with her privacy because they frequently ask her to be a babysitter for the grandchildren. This situation is quite inconvenient for her, as she has many other issues to deal with during the week. What should she do? (Artistico et al., 2003).
Let’s say that a woman has just been widowed and lives alone. What can she do to continue associating with people? (Denney & Pearce, 1989; Marsiske & Willis, 1995).
A person who lives alone wants to see her/his grandchildren more frequently. What should she/he do? (Artistico et al., 2003).

Note. *All vignettes have been previously used in adult developmental studies of EPS (e.g., Artistico et al., 2003; Denney & Palmer, 1981; Denney & Pearce, 1989; Haught et al., 2000; Haught & Walls, 2007; Marsiske & Willis, 1995) with minor wording modifications.

References

Allaire

J. C.

Marsiske

(1999). Everyday cognition: Age and intellectual ability correlates. Psychology and Aging, 14, 627–644.

Allaire

J. C.

Marsiske

(2002). Well- and ill-defined measures of everyday cognition: Relationship to older adults’ intellectual ability and functional status. Psychology and Aging, 17, 101–115.

Allaire

J. C.

Willis

S. L.

(2006). Competence in everyday activities as a predictor of cognitive risk and mortality. Aging, Neuropsychology and Cognition, 13, 207–224.

Artistico

Cervone

Pezzuti

(2003). Perceived self-efficacy and everyday problem solving among young and older adults. Psychology and Aging, 18, 68–79.

Baltes

P. B.

(1993). The aging mind: Potential and limits. The Gerontologist, 33, 580–594.

Berg

C.A.

Meegan

S. P.

Klaczynski

(1999). Age and experiential differences in strategy generation and information requests for solving everyday problems. International Journal of Behavioral Development, 23, 615–639.

Berg

C. A.

Strough

J. L.

Calderone

K. S.

Sansone

Weir

(1998). The role of problem definitions in understanding age and context effects on strategies for solving everyday problems. Psychology and Aging, 13, 29–44.

Blanchard-Fields

(2007). Everyday problem solving and emotion: An adult developmental perspective. Current Directions in Psychological Science, 16, 26–31.

Blanchard-Fields

Mienaltowski

Seay

R. B.

(2007). Age differences in everyday problem-solving effectiveness: Older adults select more effective strategies for interpersonal problems. Journal of Gerontology: Psychological Sciences, 62B, P61–P64.

10.

Carstensen

L. L.

Mikels

J. A.

(2005). At the intersection of emotion and cognition: Aging and the positivity effect. Current Directions in Psychological Science, 14, 117–121.

11.

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155–159.

12.

Cornelius

S. W.

Caspi

, (1987). Everyday problem solving in adulthood and old age. Psychology and Aging, 2, 144–153.

13.

Crawford

Channon

(2002). Dissociation between performance on abstract tests of executive function and problem solving in real-life-type situations in normal aging. Aging and Mental Health, 6, 12–21.

14.

Delis

D. C.

Kaplan

Kramer

J. H.

(2001). Delis–Kaplan executive function system. San Antonio, TX: The Psychological Corporation.

15.

Delis

D. C.

Kramer

J. H.

Kaplan

Ober

(2000). The California verbal learning test—adult version (2nd ed.). San Antonio, TX: The Psychological Corporation.

16.

Denney

N. W.

(1990). Adult age differences in traditional and practical problem solving. In Lovelace

E. A.

(Ed.), Aging and cognition: Mental processes, self-awareness, and interventions (pp. 329–349). Oxford, UK: North-Holland.

17.

Denney

N. W.

Palmer

A. M.

(1981). Adult age differences on traditional and practical problem-solving measures. Journal of Gerontology, 36, 323–328.

18.

Denney

N. W.

Pearce

K. A.

(1989). A developmental study of practical problem solving in adults. Psychology and Aging, 4, 438–442.

19.

Denney

N. W.

Pearce

K. A.

Palmer

A. M.

(1982). A developmental study of adults’ performance on traditional and practical problem-solving tasks. Experimental Aging Research, 8, 115–118.

20.

Ekstrom

R. B.

Harmon

Derman

(1976). Manual for kit of factor-referenced cognitive tests. Princeton, NJ: Educational Testing Services.

21.

Gelb

S. R.

Shapiro

R. J.

Thornton

W. J. L.

(2010). Predicting medication adherence and employment status following kidney transplant: The relative utility of traditional and everyday cognitive approaches. Neuropsychology, 24, 514–526.

22.

Gilhooly

M. L.

Gilhooly

K. J.

Phillips

L. H.

Harvey

Brady

Hanlon

(2007). Real-world problem solving and quality of life in older people. British Journal of Health Psychology, 12, 587–600.

23.

Haught

P. A.

Hill

L. A.

Nardi

A. H.

Walls

R. T.

(2000). Perceived ability and level of education as predictors of traditional and practical adult problem solving. Experimental Aging Research, 26, 89–101.

24.

Haught

P. A.

Walls

R. T.

(2007). Longitudinal adult practical problem-solving: 1983, 1993, 2003. Educational Gerontology, 33, 395–419.

25.

Heidrich

S. M.

Denney

N. W.

(1994). Does social problem solving differ from other types of problem solving during the adult years? Experimental Aging Research, 20, 105–126.

26.

Kirasic

K. C.

Allen

G. L.

Dobson

S. H.

Binder

K. S.

(1996). Aging, cognitive resources, and declarative learning. Psychology and Aging, 11, 658–670.

27.

Marsiske

Margrett

J. A.

(2006). Everyday problem solving and decision making. In Birren

Schaie

(Eds.), Handbook of the psychology of aging (6th ed., pp. 315–342). Amsterdam, Netherlands.

28.

Marsiske

Willis

S. L.

(1995). Dimensionality of everyday problem solving in older adults. Psychology and Aging, 10, 269–283.

29.

Miyake

Friedman

N. P.

Emerson

M. J.

Witzki

A. H.

Howerter

(2000). The unity and diversity of executive functions and their contributions to complex “frontal lobe” tasks: A latent variable analysis. Cognitive Psychology, 41, 49–100.

30.

Strough

McFall

J. P.

Flinn

J. A.

Schuller

K. L.

(2008). Collaborative everyday problem solving among same-gender friends in early and late adulthood. Psychology and Aging, 23, 517–530.

31.

Thornton

A. E.

Kristinsson

DeFreitas

V. G.

Thornton

W. J. L.

(2010). The ecological validity of everyday cognition in hospitalized patients with serious mental illness. Journal of Clinical and Experimental Neuropsychology, 32, 299–308.

32.

Thornton

W. J. L.

Deria

Gelb

Hill

Shapiro

R. J.

(2007). Neuropsychological mediators of the links among age, chronic illness, and everyday problem solving. Psychology and Aging, 22, 470–481.

33.

Thornton

W. J. L.

Dumke

H. A.

(2005). Age differences in everyday problem-solving and decision-making effectiveness: A meta-analytic review. Psychology and Aging, 20, 85–99.

34.

Weatherbee

S. R.

Allaire

J. C.

(2008). Everyday cognition and mortality: Performance differences and predictive utility of the everyday cognition battery. Psychology and Aging, 23, 216–221.