WIXLIB

N. Mahendra, S. Arkin / Journal of Communication Disorders 36 (2003) 395–422399characteristics to the other 20 participants. No significant differences were foundbetween the two groups in age, years of education, mean MMSE and ABCDscores, stage of dementia as measured by the Clinical Dementia Rating (CDR)(Berg, 1988), or number of comorbidities.Participants were diagnosed, tested, and enrolled according to criteria, procedures, and instruments used by the Consortium to Establish a Registry forAlzheimer’s Disease (CERAD; Morris et al., 1989). All met the criteria for aclinical diagnosis of Alzheimer’s disease as established by the National Instituteof Neurological and Communicative Disorders and Stroke/Alzheimer’s Diseaseand Related Disorders Association (NINCDS/ADRDA; McKhann et al., 1984).All participants passed a speech discrimination screening administered as part ofthe ABCD.3.2. Student Rehab partnersUndergraduate student participants were recruited from many different departments in response to mass emails sent to honor students, pre-health sciencemajors, and other likely groups. Eighty percent of the participants were females.Students were accepted on a first-come, first-served basis; the only requirementsbeing access to a motor vehicle, a current driver’s license, and auto insurance. TheUniversity’s Risk Management department provided liability insurance coveragesecondary to the student’s own insurance. In 4 years of student-providedtransportation and supervision of study participants, no accidents or injuriesoccurred during program activities. The students earned three independent studycredits for one semester of participation. Their responsibilities included drivingthe patient to and from program activities in their (the student’s) personal vehicle;supervising one exercise plus memory and language stimulation session, and onecommunity activity plus walking session per week for 10 weeks. Additionally,each student attended six, 2-h seminars related to dementia, exercise, and agingand received group and individual training and supervision in the administrationand recording of intervention activities.3.3. Cognitive–linguistic assessmentStudy participants were assessed at enrollment and after every two semestersof intervention with a battery of standardized cognitive and language tests, plusthe battery of discourse measures created for this study. The standardized testsused were the CERAD neuropsychological battery which includes the MMSE, a15-item short version of the Boston Naming test (BNT; Kaplan, Goodglass, &Weintraub, 1983), and a timed category fluency test; the ABCD; and the WAIS-RSimilarities and Comprehension sub-tests (Wechsler, 1981). Study participantswere also assessed by a CERAD-certified clinician according to the CDR, astructured interview protocol used by the CERAD to establish the stage ofdementia (Morris et al., 1989).

400N. Mahendra, S. Arkin / Journal of Communication Disorders 36 (2003) 395–4223.4. Discourse outcome measuresThe discourse battery used to monitor language performance of Elder Rehabparticipants had three parts: (a) eight stimulus prompts or questions on a variety oftopics relevant to mature adults, (b) a five-item proverb interpretation (PI) task,and (c) a picture description task. The tasks were selected to elicit the fivedifferent discourse types described by Shadden (1995). (For a detailed discussionof the rationale for each component, see Arkin & Mahendra, 2001a.)(A) Eight stimulus prompts:1. Tell me what you know about John F. Kennedy and his family.2. Tell me what you know about Alzheimer’s disease.3. This was used to assess insight as well as discourse (Arkin &Mahendra, 2001b, in press).4. Tell me about your daily activities, the things you do every day.5. Tell me about the things you do once in a while, not every day.Prompts 3 and 4 assessed participant discourse and level of awarenessabout their current life.6. Tell me the childhood thoughts and memories the word ‘‘play’’reminds you of.7. Tell me the adult thoughts and memories the word ‘‘play’’ remindsyou of.8. How would you go about planning a picnic for your family or somefriends?9. Supposing the 13-year-old daughter of a neighbor told you she waspregnant, but was afraid to tell her mother. What would you do? Whatare some of the ways the family could handle that situation? Thesecond part of Prompt 8 was designed to see how many solutionsrespondents could suggest for the presented situation.Responses to these prompts were audio and videotaped, and transcribed by students. Transcripts were demarcated into utterances andcoded, according to procedures previously described (Arkin & Mahendra, 2001a). Responses to each prompt were scored in terms of: (a) totalutterances; (b) topic comments (complete, content-filled, non-ambiguous, ‘‘non-empty’’ statements relevant to the topic under discussion);(c) neutral utterances; (d) negative utterances (e.g., confabulations,repetitions, incomplete thoughts, meaning unclear statements); (e) totalnouns; (f) different nouns; and (g) vague nouns (e.g., stuff, thing). Ratioswere computed to determine percentage of positive and negative utterances and different noun types relative to total utterances. High interand intra-scorer agreement ( 80%) on utterance demarcation, classification, and noun counts was established after extensive training anddiscussion on instances of disagreement (see Arkin & Mahendra, 2001afor details).

N. Mahendra, S. Arkin / Journal of Communication Disorders 36 (2003) 395–422401(B) Five-item proverb interpretation task: The five-item proverb interpretation task was adapted from Delis, Kramer, and Kaplan (1984). The fiveproverb stimuli used were:1. They see eye to eye.2. Too many cooks spoil the broth.3. Rome wasn’t built in a day.4. Don’t count your chickens before they hatch.5. You can’t tell a book by its cover.Participant responses were scored along a seven-point concreteness–abstractness continuum (Chapman et al., 1997—Appendix B) with 6being a complete abstract response and 0 being an incorrect or noresponse. The highest possible score on the test was 30 (5 items 6points). To ensure reliability of scoring, baseline and first year postintervention transcripts were scored by Chapman and colleagues, usingtheir guidelines (Chapman et al., 1997). At the end of the first year, thesecond author scored all patient responses, following Chapman’s guidelines and consulting with her on questionable responses. Intra-scoreragreement was obtained after re-scoring 25% of participant transcriptsand determining item-by-item reliability to be 75%.(C) Grocery store picture description: This task is from the AphasiaDiagnostic Profiles (ADP; Helm-Estabrooks, 1992). Participants wereasked to describe the picture in as much detail as possible and when theystopped, they were encouraged to provide more information by use ofgeneral prompts such as ‘‘Is there anything else you can tell me aboutthis picture?’’ Descriptive discourse elicited was scored according tothe guidelines and list of acceptable sample information units providedin the ADP manual. Prepositional phrases, multiple verb forms, aswell as pronouns relating to a new topic (personal communication,Helm-Estabrooks, 1997) were also given credit.4. InterventionsStudent participants administered one 2–2.5-h exercise plus language stimulation session, and supervised one 1.5–2-h community activity session per week for10 weeks per semester. A family member supervised 10 weekly exercise sessions.4.1. Physical fitnessThe physical fitness intervention that provided the context for the language andmemory stimulation activities consisted of a treadmill and stationary bicycleworkout, strength training on five machines, and flexibility and balance exercises.Details and outcomes have been described elsewhere (Arkin, 1999, in press).

402N. Mahendra, S. Arkin / Journal of Communication Disorders 36 (2003) 395–4224.2. Community activity sessionsSessions alternated between supervised volunteer service at communityagencies and cultural or recreational activities (Arkin, 2001). Students wereinstructed to incorporate 20 min of walking into the community activity sessions,to bring the frequency of aerobic activity to the usually recommended three timesper week.4.3. Cognitive–linguistic activitiesThe portfolio of language activities provided practice in the importantcognitive operations identified by Orange and Purves (1996) as relevant toconversational performance. Each activity and the cognitive process/es it tappedinto are listed below. Students were instructed and repeatedly reminded to usethese exercises as a launchpad for dialogue between them and their partners, notsimply to record a first response and quickly move on. Each activity was labeledeither ‘‘during’’ or ‘‘rest’’ to indicate whether it was to be administered during anaerobic activity or during a rest period. Tasks requiring concentration or deepprocessing were reserved for rest periods. All activities required participants tomaintain attention to the task and to perform some form of deep processing (Craik& Lockhart, 1972) in response to a presented stimulus and to produce verbalresponses. A complete list, brief description, and examples of language activitiesused in the Elder Rehab program are presented in Table 2. Nine or 10 were usedduring any given semester. Language session report booklets containing participant responses to the various activities were turned in weekly by the students.Record cards containing aerobic and strength training achievements were alsocompleted for each session.5. ResultsThe effects of the interventions on the four participants’ performance onpreviously listed discourse outcome measures are reported below.5.1. Topic comment to total utterance (TC/U) ratioBecause our interest was in increasing appropriate and on-topic verbalutterances, the chief outcome measure was a ratio of topic comments to totalutterances. These data are presented in Fig. 1. Most noteworthy is that two of thefour (S1 and S4) participants improved their ratios from baseline to year 4. S1’simprovement was modest and consistent, except for a leveling off between years 3and 4. S4’s was erratic, with substantial dips in years 2 and 3 and a dramaticincrease in year 4. The two non-improvers (S2 and S3) also showed declines intheir second and third years, but S2 again showed improvement in her fourth year.

N. Mahendra, S. Arkin / Journal of Communication Disorders 36 (2003) 395–422403Table 2Language activities used in the Elder Rehab programPicture descriptionRequires attention, visual scanning, naming, narrative discourse production,and processing of literal and interpretive concepts in the picture (e.g., in theGrocery store picture task, a literal concept is ‘‘a child spilling eggs’’ and aninterpretive concept is ‘‘no one is paying attention to the child’’). Advantagesof picture description tasks for persons with dementia are that this does notinvolve memory and concepts missed can be pointed out by directing theparticipants’ attention to the details or contextual cues in the picture. Theprogram’s use of this activity differed from the use of picture description asan assessment instrument, when only non-specific cues are given. Thisactivity typically took place at the beginning of the session. A NormanRockwell was presented. Students would prompt: ‘‘Tell me what you see inthat picture—what’s going on?’’ After recording the free response, studentsasked questions (from a prepared list in a resource manual) designed tofocus attention on details and relationships and to elicit inferences frompictorial cuesCar bingoRequires orientation, attention, reading words and short phrases, visuospatialscanning (to search for objects), and procedural memory for what a ‘‘bingo’’activity involves. This activity occurred during the car ride to and from theintervention site. A large bingo-type card was prepared, each squarecontaining the name of an object that might be seen from a car window (e.g.,red pickup truck, driver wearing a hat, dead animal, school bus). Participantswere challenged to look for one or more items at a time during a given trip,depending on their ability to focus on more than one item at a time. The dateon which an item was observed was written on the card’s appropriate squareObject descriptionUtilizes lexical and semantic memory and requires an individual to generatemeaningful attributes of a common object. Three objects (lemon, pencil,toothbrush) were presented in order and participants were asked to describethem. Based on the spontaneous response, prompt questions were asked toelicit additional informationStory recallwith quizThis activity employs a fixed interval type of spaced retrieval (Camp et al.,1995) and challenges participants’ explicit and episodic memory. The formatis similar to that used in a series of successful autobiographical memorytraining studies with Alzheimer patients (Arkin, 1992, 1997, 1998, 2000).Students read a brief story containing six–seven facts and asked partners tore-tell the story immediately after presentation. Students then re-read story,posing a question about each fact after it was stated, pausing for the partnerto answer, if able, then giving the correct answer. Students then repeated allsix questions and solicited their partner’s answers, giving the correct one if s/he failed to answer correctly. The partner was then asked to re-tell the story.When five out of six questions were answered correctly and the partnerspontaneously recalled four major facts about the story, a new story wasintroduced for the following session. Topics of stories were adult in contentand emotionally charged (see Appendix C for an example)

404N. Mahendra, S. Arkin / Journal of Communication Disorders 36 (2003) 395–422Table 2 (Continued )Category fluencynaming-quizUtilizes the relatively preserved implicit memory systems of AD patients(Bäckman, 1992; Camp et al., 1995) and fosters production of non-studiednovel exemplars through the process of spreading activation (Bayles &Kaszniak, 1987; Collins & Loftus, 1975). A 60-s category fluency test wasadministered (e.g., things people wear) 18–20 times in a given semester,typically before and after each exercise session. A master (baseline) list of allthe different exemplars ever named by each participant during those tests wascompiled. The subsequent semester, participants were presented with a 40- to50-item study task at the start of their exercise session. Pictures of items fromthe subject category were alternated with questions about the same topic(e.g., What might a woman wear to bed instead of pajamas?). Correctanswers were supplied. At the end of the exercise session, the categoryfluency test was given again. Lists produced following exposure to studytasks were analyzed for (a) words never named at baseline, but that were onthe study task—evidence for implicit learning; and (b) novel words, i.e.,words never named at baseline and that were not on the study task—evidenceof spreading activation. Results obtained for the clothing category on thistask are published elsewhere (Arkin, Rose, & Hopper, 2000)Opinion andadvice questionsRequires participants to use logical thinking, reasoning, and moral judgment.A situation involving a moral issue or personal values is presented toparticipants and their opinion solicited. Probe questions were asked afterrecording an initial, spontaneous response. Controversial adult topics anddilemmas were presented. For example, ‘‘Would you help a terminally illfriend hide pills he was accumulating in order to commit suicide?’’ Thisactivity was successfully used in a geropsychiatric hospital in Germany tocounteract loss of social roles and functions that accompanies institutionalization (Muller, 1993)Pros and consEnables participants to practice reasoning skills. Participants were given aseries of topics or objects and were asked to tell what was good and bad abouteach one. Students used topics and responses as a springboard for conversation,sharing their opinions during the dialogue. For example, retirement, cellphones, etc.Problem solvingFocuses on participants’ critical thinking and reasoning skills. One of aseries of picture cards developed for use with adults with disabilities (Pitti &Meier, 1992) was presented at each session. Each card depicts a dailyliving situation in which a problematic or dangerous element is present. Forexample, a depicted situation might include someone reaching into an ovenwith bare handsA my nameis . . . gameProvides practice in generative naming, set shifting, and executive functionskills. Students read the framework phrase and asked participants to producean appropriate word (man’s/woman’s name, place name, etc.) beginning withand my (wife’s) namethe target letter for each blank (A my name isisand we come fromand we’re going downtown to buy;B, etc.). Higher functioning participants were asked to advancemy name isthrough the alphabet with each word produced (e.g., A: Alice, B: Bob,C: Chicago, etc.)

N. Mahendra, S. Arkin / Journal of Communication Disorders 36 (2003) 395–422405Table 2 (Continued )SimilaritiesTargets abstract reasoning skills and is similar to that tested by the WAIS-Rsubtest. Pairs of words belonging to the same superordinate category werepresented and participants were asked to name the category. They were askedwhat the two words in each pair of items had in common—how they werealike. If they named an attribute or a function, they were corrected and toldthe correct category name and were asked to repeat it. For example, Cup andplate(both dishes) (not, ‘‘you eat from both’’)Proverb completionand interpretationEngages abstract verbal reasoning (Lezak, 1995) and higher order cognitiveprocesses of figurative language comprehension (Van Lancker, 1990).Students read beginnings of proverbs to see if participants knew the ending.If not, students gave the ending. Then students asked participants what theproverb meant. If participants gave a concrete or incorrect response, studentsgave the abstract interpretation. Then proverbs were read again andparticipants asked again for their interpretation. For example, A bird in the(is worth two in the bush). Proverbs and their interpretationshandwere taken from The Dictionary of Cultural Literacy (Hirsch, Kett, & Trefil,1988). This activity was difficult for many patients, and some students. Aneasier, often more successful variant was to read beginning lines of a list ofproverbs and have subjects provide endings. After a few times, participantstypically learned the correct responsesFamous namesStimulates recognition memory for the names of famous people. Participantswere read a series of first names of famous people and were asked to providetheir last names. Students asked participants if they could tell somethingabout that person. Identifying information for most commonly namedindividuals was supplied to the students, as they frequently did not know whosome of the famous individuals were. This was an area where participantscould often educate their student partnersWord associationProvides information on lexical–semantic organization by triggering retrievalof concepts and ideas related to a target stimulus. Participants were enc

supervisors of non-professional rehab partners for persons with dementia—is proposed. . ASHA's Code of Ethics and the OBRA (Omnibus Budget Reconciliation Act, . and auto insurance. The University's Risk Management department provided liability insurance coverage secondary to the student's own insurance. In 4 years of student-provided