Ecological validity reconsidered: the Night Out Task versus the D-KEFS

Conceptualization of the term "ecological validity" in neuropsychological research on executive function assessment: a systematic review and call to action

OBJECTIVE

"Ecological validity" (EV) is classically defined as test's ability to predict real-world functioning, either alone or together with test's similarity to real-world tasks. In neuropsychological literature on assessment of executive functions (EF), EV is conceptualized inconsistently, leading to misconceptions about the utility of tests. The goal of this systematic review was to examine how EV is conceptualized in studies of EF tests described as ecologically valid.

METHOD

MEDLINE and PsychINFO Databases were searched. PRISMA guidelines were observed. After applying inclusion and exclusion criteria, this search yielded 90 articles. Deductive content analysis was employed to determine how the term EV was used.

RESULTS

About 1/3 of the studies conceptualized EV as the test's ability to predict functional outcomes, 1/3 as both the ability to predict functional outcome and similarity to real-world tasks, and 1/3 were either unclear about the meaning of the term or relied on notions unrelated to classical definitions (e.g., similarity to real-world tasks alone, association with other tests, or the ability to discriminate between populations).

CONCLUSIONS

Conceptualizations of the term EV in literature on EF assessment vary grossly, subsuming the notions of criterion, construct, and face validity, as well as sensitivity/specificity. Such inconsistency makes it difficult to interpret clinical utility of tests that are described as ecologically valid. We call on the field to require that, at minimum, the term EV be clearly defined in all publications, or replaced with more concrete terminology (e.g., criterion validity).

Computerized assessment of executive functioning: Validation of the CNS Vital Signs executive functioning scores in a sample of community-dwelling older adults

Objective: Computerized assessment of cognitive functioning has gained significant popularity over recent years, yet options for clinical assessment of executive functioning (EF) are lacking. One computerized testing platform, CNS Vital Signs (CNS-VS), offers tests designed to measure EF but requires further validation. The goal of the present study was to validate CNS-VS executive scores against standard clinical measures of EF. We also sought to determine whether a modified CNS-VS composite score that included variables purported to measure inhibition, switching, and working memory would outperform the currently available CNS-VS Executive Function Index. Method: A sample of 73 cognitively healthy older adults completed four tests from the Delis-Kaplan Executive Function System, the Digit Span subtest from the Wechsler Adult Intelligence Scale-fourth edition, and three CNS-VS tasks purported to measure inhibition, switching, and working memory. Results: Performances on the CNS-VS tests were predicted by performances on standard paper-and-pencil measures. Although the currently available CNS-VS Executive Function Index predicted unique variance in a well-validated paper-and-pencil EF composite score, our Modified CNS-VS EF composite accounted for unique variance above and beyond the original CNS-VS Executive Function Index, while the reverse was not true. Conclusions: The present results support the construct validity of CNS-VS EF tests but also suggest that modifications to their current composite scores would improve the prediction of EF performance.

Six elements test vs D-KEFS: what does "Ecological Validity" tell us?

OBJECTIVE

Extensive research shows that tests of executive functioning (EF) predict instrumental activities of daily living (IADLs) but are nevertheless often criticized for having poor ecological validity. The Modified Six Elements Test (MSET) is a pencil-and-paper test that was developed to mimic the demands of daily life, with the assumption that this would result in a more ecologically valid test. Although the MSET has been extensively validated in its ability to capture cognitive deficits in various populations, support for its ability to predict functioning in daily life is mixed. This study aimed to examine the MSET's ability to predict IADLs assessed via three different modalities relative to traditional EF measures.

METHOD

Participants (93 adults aged 60 - 85) completed the MSET, traditional measures of EF (Delis-Kaplan Executive Function System; D-KEFS), and self-reported and performance-based IADLs in the lab. Participants then completed three weeks of IADL tasks at home, using the Daily Assessment of Independent Living and Executive Skills (DAILIES) protocol.

RESULTS

The MSET predicted only IADLs completed at home, while the D-KEFS predicted IADLs across all three modalities. Further, the D-KEFS predicted home-based IADLs beyond the MSET when pitted against each other, whereas the MSET did not contribute beyond the D-KEFS.

CONCLUSIONS

Traditional EF tests (D-KEFS) appear to be superior to the MSET in predicting IADLs in community-dwelling older adults. The present results argue against replacing traditional measures with the MSET when addressing functional independence of generally high-functioning and cognitive healthy older adult patients.

Test of Practical Judgment (TOP-J): Construct, Criterion, and Incremental Validity in a Community Sample of Older Adults

OBJECTIVE

The Test of Practical Judgment (TOP-J) is a stand-alone judgment measure that is considered to tap into aspects of executive functioning (EF) and inform clinical predictions of daily functioning in older adults. Past validation research is variable and has some limitations. The present study sought to examine the reliability and construct, criterion, and incremental validities of scores on TOP-J 9-item version (TOP-J/9).

METHOD

Participants were 95 community-dwelling older adults aged 60 to 85. Participants completed TOP-J/9, measures of EF and global cognition, and three different modalities of instrumental activities of daily living (IADLs) (self-report, performance-based tasks conducted in the laboratory, and performance-based tasks completed at home over 3 weeks).

RESULTS

TOP-J/9 scores showed adequate internal consistency (α = 0.73) after correcting for the low number of items. TOP-J/9 was correlated with global cognition and EF, although EF did not survive correction for lower-order processes. Finally, although TOP-J/9 scores were associated with home-based IADL tasks (but not with self-report and laboratory-based IADLs), providing some evidence of criterion validity, they did not incrementally contribute to home-based IADL performance beyond other cognitive measures. However, when two items pertaining to social/ethical judgment were removed, this modified version of TOP-J did relate to EF beyond lower-order processes and contributed uniquely to prediction of home-based IADLs beyond other measures.

CONCLUSION

Results suggest that TOP-J/9 taps into global cognitive status (but not necessarily EF) and predicts "real-world" functioning (but not above and beyond other cognitive measures). TOP-J psychometrics may be improved by removing two social/ethical items.

Neuropsychological assessment methodology revisited: metatheoretical reflections

Theory building in neuropsychology, similar to other disciplines, rests on metatheoretical assumptions of philosophical origin. Such assumptions regarding the relation of psychological and physiological variables influence research methodologies as well as assessment strategies in fields of application. Here, we revisit the classic procedure of Double Dissociation (DD) to illustrate the connection of metatheory and methodology. In a seemingly unbridgeable opposition, the classical neuropsychological procedure of DD can be understood as either presupposing localizationism and a modular view of the brain, or as a special case of the generalized neuro-lens model for neuropsychological assessment. In the latter case, it is more easily compatible with a perspective that emphasizes the systemic-network, rather than the modular, nature of the brain, which as part of the organism, proportionately mediates the situatedness of the human being in the world. This perspective not only makes it possible to structure ecological validation processes and give them a metatheoretical foundation, but also to interlace it with the phenomenological insight that the laboratory as one context of empirical research may be analyzed in terms of situated experience. We conclude with showing that both the localizationist and the system science approach can agree on a view of the brain as a dynamical network, and that metatheory may thus offer important new perspectives of reconciliation.