Art & science research Effect of time of day on language in and Studies into how circadian rhythms affect patients abilities to engage with staff will help inform practice in the future, say Amanda Stead and colleagues Correspondence stea3478@pacificu.edu Amanda Stead is assistant professor, Pacific University, School of Communication Sciences and Disorders, Oregon, US Neila Donovan is associate professor, communication sciences and disorders and Paul Hoffman is professor, communication sciences and disorders Both at Louisiana State University, Louisiana, US Date of submission December 11 2014 Date of acceptance February 19 2015 Peer review This article has been subject to double-blind review and has been checked using antiplagiarism software Author guidelines journals.rcni.com/r/nop-authorguidelines Abstract Aim To investigate whether narrative discourse followed a diurnal pattern across one ten-hour day in people and those with mild to moderate (AD). Method Ten people and ten clinically labelled with probable AD were recruited. Measurements of language and cognition were collected across one day at 9am, 12pm, 3pm and 6pm. Language samples were evaluated for quantity (total utterances and words per minute) and quality: mazes or fillers, repetitions and revisions, abandoned utterances and type token ratio or percentage of different words to total words. Results The group performed significantly better on cognitive measures across the FOR HEALTHCARE professionals who work with people with (AD), identifying factors that contribute to or detract from successful communication is imperative to ensure proper care and quality of life. To achieve that goal, researchers need a way to differentiate between cognitive and language changes that occur due to healthy ageing processes and those due to AD. Identifying these differences could result in more appropriate assessment and treatment for people with AD. Information from narrative discourse The use of conversations or stories, that is, narrative discourse, is a primary means of communicating personal information, which provides an opportunity day than the AD group. At all times the group produced significantly longer narrative samples that were significantly less aborted and revised than the AD group. Additionally, both groups demonstrated declining narrative performance as the day progressed. Conclusion Based on these results, time of day may be an additional factor that moderates narrative performance. This change in narrative ability may have an effect on making a proper diagnosis, therapeutic effectiveness and patient interactions, therefore affecting quality of care. Keywords Ageing,, circadian, cognition, dementia, language to investigate if other factors affect performance or ability. Previous research has shown that people with AD carry on conversations differently from healthy older people. People with AD: Have less information content propositional (Ripich et al 2000, Groves-Wright et al 2004, Carlomagno et al 2005). Produce sentences that are not appropriately related to the topic cohesion and coherence (Glosser and Deser 1991, Dijkstra et al 2004). Produce sentences that are shorter in length (Bowles and Poon 1985, Drevenstedt and Bellezza 1993, Kemper et al 2001, Bayles and Tomoeda 2007). NURSING OLDER PEOPLE April 2015 Volume 27 Number 3 31
Art & science research Additionally, researchers investigating discourse, larger units of speech, in people with AD and mild cognitive impairment have reported deficits including: Poor topic maintenance (Ulatowska and Chapman 1995, Guendouzi and Müller 2008). Poor conversational repair (Orange et al 1996). Higher usage of indefinite pronouns (Santo Pietro and Berman 1984, Ulatowska et al 1988). A limited vocabulary (Tomoeda et al 1996, Bucks et al 2000, Guendouzi and Müller 2008). The current study explored spoken narrative discourse, the recounting of an event or series of events in chronological order. Narratives are an important aspect of communication because they are often used to tell something about one s self or about an event. They are essential for social communication (Guendouzi and Müller 2008). Narratives are not only socially essential, but are also used frequently in the retelling of medical history and events, which is vital for quality of care. Through the use of narrative, people are able to communicate wants and needs, however, narrative discourse functions only when its underlying cognitive processes are intact: memory, attention and planning. Any process that impairs cognition is also likely to impair language discourse, thus affecting one s ability to communicate effectively. Because AD impairs cognition so significantly, narrative is impaired even in the early stages of the disease. Circadian rhythms Additional factors other than cognitive status can influence performance. The underlying cognitive processes that are affected in AD, such as memory and attention, are known to be affected by changes in circadian rhythms (CRs). CRs are the patterns of arousal regulated by one s internal clock. Changes in CRs are common in older adults and variations in morning and evening preference and performance have been examined across the lifespan (May et al 1993). CR research has focused on diurnal changes in memory, attention and inhibition as a function of age. This research suggests that during times of self-reported peak arousal people have improved performance on complex tasks (Winocur and Hasher 1999, Hasher et al 2002, May et al 2005). When people perform a complex task during a non-peak time their performance is worse and this is further affected by the presence of AD. This could mean that the same person may appear better or worse depending on the time of day they are tested. This is particularly important when it comes to the diagnosis and medication management of people with AD. Language discourse processing A limited number of studies have examined language discourse processing or production for diurnal patterns (Folkard 1975, Oakhill 1986a, 1986b, 1988, Reinberg et al 1988, Morton and Diubaldo 1995, Yaretsky et al 1995, 1996). These studies report variable findings where some aspects of language, for example, sentence comprehension, text processing and word fluency, may be vulnerable to changes in CRs, while others may not, for example, syllable perception. Determining if language discourse is affected by changes in arousal is particularly important since the expression of ideas and verbal performance for assessment is essential for quality of care. The relationship between cognitive status and arousal, and their effect on narrative discourse were the motivation for this article. Aim The aim of this study was to investigate whether narrative discourse followed a diurnal pattern, as does memory and attention, across one ten-hour day in individuals and those with mild to moderate AD. Three experimental questions were asked and a hypothesis was given for each question: 1. Was there a significant relationship between measures of cognition and narrative discourse across four times in one day? 2. Was there a significant difference between the quality of the two groups narrative discourse? 3. Was there a significant difference between the quantity of the two groups narrative discourse? It was hypothesised that: 1. A relationship would exist between measures of cognition and narrative discourse across the day. 2. Measures of narrative discourse quality would decrease across the day for both groups. 3. Measures of narrative discourse quantity would decline across the day for both groups. Method This study used a prospective, quasi-experimental design measuring the quality and quantity of narrative discourse in two groups. Twenty participants were recruited from surrounding communities through the local Alzheimer s respite centre and through word of mouth. All participants: Were native English speakers. Had no history of depression or major psychiatric disorders. Had at least eight years of formal education. Passed a hearing screening (aided). Group 1 consisted of ten adults aged 65-89 years. All participants in this group scored 32 April 2015 Volume 27 Number 3 NURSING OLDER PEOPLE
28 or higher on the mini mental state examination (MMSE) to exclude the presence of cognitive dysfunction (Folstein et al 1975). Group 2 consisted of ten adults aged 65-89 clinically labelled as having probable AD. They scored between 11 and 25 on the MMSE. All participants expressed a preference for morning based on the morningness-eveningness questionnaire (Horne and Ostberg 1977). Before the study the researcher assessed participants in their home over two days. On the first day all eligibility examinations were given. Participants gave consent, were asked to fill out a health status questionnaire, were given the MMSE, passed a vision and hearing screening and passed a speech discrimination task from the Arizona Battery for Communication Disorders of Dementia (Bayles and Tomoeda 1993). On day two, the researcher collected language samples at four points across the day: at around 9am, 12pm, 3pm and 6pm. One narrative question from a corpus of four common probes was asked during each visit. These questions involved asking the participant about their: day, family, last vacation and career. The four questions were randomised across visits and participants using an online randomiser (Social Psychology Network 1997-2015). In addition to the narrative language sample taken at each time period, participants were assessed for cognitive status using the Benton Judgment of Line Orientation Test (BJLO) (Benton et al 1978). The BJLO was chosen to assess cognitive status at the time of sample elicitation because it uses relatively little verbal language. Language samples were typed and coded into Systematic Analysis of Language Transcripts (SALT) (Miller and Chapman 2000). SALT is analytical software used to evaluate language transcripts for linguistic variables. These variables examined in the present study included: Total utterances: the total count of utterances produced during the structured task; that is, an independent clause and its modifiers. Abandoned utterances: utterances in which the speaker stopped mid-utterance resulting in a sentence fragment. Type token ratio (TTR): the ratio of different words (types) to total words (tokens). Words per minute (WPM): the number of WPM of sample. Percentage of mazes used: percentage of maze words used as a percentage of total words. Maze words include fillers, for example, uh, repetitions and revisions. Ethical considerations The Louisiana State University institutional review board approved the study and all participants provided informed consent before any data collection. Primary carers were consented in the case of a participant with dementia, and participants assented. Results All statistical analysis was conducted using Stata, version 12. A generalised linear model for each dependent variable was specified with two levels of group ( (HA)=0, AD=1) and four levels of time (9am, 12pm, 3pm and 6pm). In terms of group equality, the average ages of the two groups were not statistically different (t (28)=0.9762, P<0.3419), nor was the gender composition (χ2 (1)=0.6392, P<0.6392). As expected, the HA group achieved MMSE scores that were significantly higher than those of the AD group (t(18)=10.182, P<0.0001). Research question 1 It was hypothesised that a relationship would exist between measures of cognition and discourse across the day. This was not found and no correlations reached statistical significance. However, the BJLO showed that the HA group performed significantly better than the AD group (z=-1.97, P=0.049). Also, the interaction was significant between 9am and 6pm (z=-2.05, P=0.041), indicating that the BJLO scores of both groups were falling but at different rates. Figure 1 (page 34) shows the average group scores as a function of time of day with associated standard error of estimates. The AD group showed a more rapid decline extending across the whole day. Research question 2 It was hypothesised that measures of narrative quality, that is, abandoned utterances, percentage of mazes and TTR, would decrease across the day for both groups. This prediction was supported. The AD group produced a greater percentage of mazes than the HA group (z=2.85, P=0.004). The group-by-time interaction was significant between 9am and 12pm (z=-2.35, P=0.019). A positive indicator of discourse structure, TTR, appeared to decrease in frequency during the day. TTR showed a significant decrease over time overall (z=-3.20, P=0.001) with the greatest decrease between 9am and 12pm (z=-2.07, P=0.039) for both groups. Abandoned utterances showed a time effect (z=6.37, P=0.000) indicating an overall increase between 9am and 6pm. The difference was also significant between 9am and 3pm (z=1.96, P=0.05). The group-by-time interaction was significant between 9am and 3pm (z=-2.22, P=0.027) and 9am and 6pm (z=-2.46, P=0.014). This shows a crossover effect where people with AD were higher on abandoned utterances at 9am and NURSING OLDER PEOPLE April 2015 Volume 27 Number 3 33
Art & science research Figure 1 Results of Benton Judgment of Line Orientation Test performance across the day in people and people with with standard error bars Benton Judgment of Line Orientation Test Score 12-1 8-6 - 4-2 - 9am 12pm 3pm 6pm people were higher on abandoned utterances at 3pm and 6pm. Figure 2 shows the average scores of the two groups across time of day for abandoned utterances, percentage of utterances that were mazes and TTR. Research question 3 It was hypothesised that measures of narrative discourse quantity, that is, total utterances and WPM, would decrease across the day for both groups. WPM showed a significant time effect between 9am and 12pm (z=3.45, P=0.001). The group-by-time interaction was also significant between 9am and 12pm (z=-2.25, P=0.025). This shows that between 9am and 12pm WPM increased sharply and the largest difference between the two groups was at 12pm (Figure 3, page 36). Total utterances showed an overall group difference (z=-3.22, P=0.001) in which people with AD produced fewer utterances overall (Figure 3, page 36). See Table 1 (page 37) for a summary of the significant findings. Discussion To the authors knowledge, this is the first study to investigate how arousal patterns might affect narrative discourse across a single day. The purpose of this study was to investigate if narrative discourse followed a diurnal pattern across a single ten-hour day in people and people with AD. This result could help establish if narrative performance was being moderated by CRs, similar to other cognitive processes. While the sample size is small and the study results cannot yet be generalised to the broader AD population, the results appear to support the methodology used in the study and support the hypotheses. In research question 1 it was hypothesised that a relationship would exist between measures of cognition and discourse across the day, meaning that if one changed the other would follow. Although not correlated with language variables, the measure of cognition (BJLO score) was significantly different between groups where healthy people performed significantly better than those with AD. This was expected and consistent with the literature documenting cognitive decline in people with AD compared with peers (Kemper et al 2001, Bayles and Tomoeda 2013). Furthermore, there was a significant group-by-time interaction from 9am to 6pm indicating that time had a distinctive influence on the ability to maintain cognitive performance. The people in the AD group performed worse than the group and showed a steeper decline, with an average score of 7.2 at 9am that declined to 4.6 by 6pm (Figure 1). This result might indicate that the cognitive processes needed to perform on the BJLO, such as visual spatial processing, are disproportionately affected when a pre-existing cognitive impairment exists. As the day progressed the participants were still able to maintain a relatively even level of performance while the AD participants dropped to a 31% success rate. This is a significant finding from this study. As there is a decrease in arousal, those with a cognitive dysfunction are not able to cope as well as those who are healthy, indicating that if a pre-existing cognitive condition exists, changes in arousal may have a disproportionate effect on performance. This may be because healthy people have a sufficient cognitive reserve to cope with the 34 April 2015 Volume 27 Number 3 NURSING OLDER PEOPLE
Figure 2 Average performance across the day on measures of narrative quality (abandoned utterances, percentage of utterances mazed and type token ratio) in people and people with with standard error bars % abandoned utterances 16-14 - 12-1 8-6 - 4-2 - 9am 12pm 3pm 6pm 6 % utterances mazed 5 4 3 2 1 9am 12pm 3pm 6pm 0.62-0.6 0.58 - Type token ratio 0.56-0.54-0.52-0.5 0.48-0.46-0.44-9am 12pm 3pm 6pm NURSING OLDER PEOPLE April 2015 Volume 27 Number 3 35
Art & science xxx research Figure 3 Average performance across the day on measures of narrative quantity (total utterances and words per minute) in people and people with with standard error bars Total utterances 35-3 25-2 15-1 5-9am 12pm 3pm 6pm Words per minute 16 14 12 10 8 6 4 2 9am 12pm 3pm 6pm additional barrier, whereas those with AD do not. This is also consistent with literature in the area of cognition and CRs, indicating that decreasing arousal can have a negative effect on performance, although in this case only for those of lower cognitive status (May et al 1993, 2005). Research question 2 hypothesised that measures of narrative quality would decrease across the day for both groups. The AD group produced significantly more mazes across the day than the group. This is supported by evidence that people with AD produce more repetitions of phrases, words and revisions than people (Ripich et al 2000, Guendouzi and Müller 2008). This increased maze production may also indicate an underlying problem in the formation of phrases (Fleming 2014). It has been hypothesised that poor phrase formulation and diminished word retrieval is an expression of a problem in higher order cognitive processes (Fleming and Harris 2008). There was also a significant group-by-time interaction from 9am to 12pm, indicating that the groups were changing at different rates over the given time period. However, this effect dissolved after 12pm. Though members of the AD group self-reported highest arousal in the morning, which should have led to increased performance, their maze production indicates the opposite. As the day went on healthy people produced more mazes and became less fluent and less precise in their narrative use. Measures of TTR indicated a significant time effect for both groups; meaning time had a direct effect on the variety of words produced in narratives 36 April 2015 Volume 27 Number 3 NURSING OLDER PEOPLE
across the day. The decline in TTR over time may be the result of poorer memory performance, which is known to be affected by shifts in CRs (Yoon et al 1998, May 1999, May et al 2005). This diminishing memory ability may have made it more difficult to generate sufficient detail when retelling a narrative leading to a less diverse sample overall. Results from cognitive studies showed performance was better during peak arousal times, as it was in the current study (Yoon et al 1998, May 1999, May et al 2005). The production of abandoned utterances was significantly affected by time of day. Additionally, both groups changed at different rates across the day. Research indicates that an increase in abandoned utterances contributes to the diminished quality of discourse (Ripich et al 2000, Guendouzi and Müller 2008). The increase in abandoned utterance scores in the healthy group may substantiate reports of increased word-finding difficulty at the end of the day in older people (Bayles 1991). Like mazes, this increase might indicate problems with the formation and organisation of narrative discourse, which leads to the diminished quality of the overall message (Ripich et al 2000, Fleming and Harris 2008, Guendouzi and Müller 2008). The quality of narrative decreased across the day for all variables, which indicates that the time at which a sample is taken may affect its quality. This is significant given that the quality of verbal interaction can often affect the perception of impairment or the quality of care. In research question 3 it was hypothesised that measures of narrative discourse quantity would decline across the day for both groups. Researchers have reported that people with AD have a slower rate of speech than healthy peers (Santo Pietro and Berman 1984, Cherney and Canter 1993, Ripich et al 2000). The current study supports this; the healthy people talked faster than those with AD. Results also indicated that the change from 9am to 12pm was the most volatile and that the groups reacted differently to the shifting time. This resulted in the largest gap in performance at 12pm. The healthy group produced significantly more phrases overall than the AD group. This is consistent with previous studies that have found people with AD produce fewer utterances than healthy peers (Ripich and Terrell 1988, Cherney and Canter 1993, Ripich et al 2000). The reduction in the number of utterances may reflect the underlying memory impairment in AD that reduces the level of detail when communicating on a topic. Table 1 Summary of significant findings Variable Significant finding P value Benton Judgment of Line Orientation Test Overall group difference 0.049 Group-by-time interaction (9am-6pm) 0.041 Abandoned utterances Overall time difference 0.000 Mazes Overall group difference 0.004 Group-by-time interaction (9am-12pm) 0.019 Type token ratio Overall time difference 0.001 Time difference (9am-12pm) 0.039 Words per minute Time difference (9am-12pm) 0.001 Group-by-time interaction (9am-12pm) 0.025 Total utterances Overall group difference 0.001 Limitations include the small number of participants and the possible influence of unknown external variables, such as mealtimes, naptimes, medication schedules and exercise, on the measures, although efforts were made to account for these. Follow-up studies should include larger groups. Future studies could also include the correlation of physiological measures to linguistic ones. Conclusion All variables investigated by this study, with the exception of total utterances, were affected by time, meaning that depending on group membership, time had an influence on the performance of narrative discourse. These findings align with existing evidence that cognitive functions are affected by CRs (May et al 1993, Intons-Peterson et al 1998, Yoon et al 1998, May 1999, May et al 2005, Schmidt et al 2007). The cognitive measure used in this study, the BJLO, differentiated between groups and showed that changes in CRs affected the groups differently. People with AD produced fewer utterances and were more diverse, more aborted and less fluent. Additionally, the 9am-12pm time period appeared to be the most volatile in terms of group and time interactions. The results represent a promising area of study for researchers and have potential clinical benefit for healthcare professionals interested in time-of-day effects on people with AD. Furthermore, future studies may be particularly useful for nurses working with patients with AD who need patients at optimal performance levels to maximise their care. NURSING OLDER PEOPLE April 2015 Volume 27 Number 3 37
Art & science research Implications for practice Cognition and language may be susceptible to changes in circadian rhythms (CRs). Doctors, nurses and therapists interacting with older patients, and particularly those with, need to be aware of such fluctuations as they monitor and report patient status. People may seem to be more or less impaired depending on the time of day they are assessed or treated. Possible causes of performance change should be taken into account because they may affect diagnostic and therapeutic effectiveness. It may be beneficial for healthcare professionals to work with patients, particularly discourse-heavy interactions, during their self-reported peak arousal times, which have been identified as a time of peak performance for multiple cognitive variables. People with existing cognitive dysfunction are less able to cope with other outside influences, such as changes in CRs, and maintain performance. This means that even seemingly small factors may have a significant effect on a person s performance or communication, thus putting them at higher risk of negative outcomes. Online archive For related information, visit our online archive and search using the keywords Conflict of interest None declared References Bayles K (1991) symptoms: prevalence and order of appearance. Journal of Applied Gerontology. 10, 4, 419-430. Bayles K, Tomoeda C (1993) Arizona Battery for Communication Disorders of Dementia. Canyonlands Publishing, Tucson AZ. Bayles K, Tomoeda C (2007) Cognitive-Communication Disorders of Dementia: Definition, Diagnosis, and Treatment. Plural Publishing, San Diego CA. Bayles K, Tomoeda C (2013) Cognitive-Communication Disorders of Dementia: Definition, Diagnosis, and Treatment. Second edition. Plural Publishing, San Diego CA. Benton A, Varney N, Hamsher K (1978) Visuospatial judgment. A clinical test. Archives of Neurology. 35, 6, 364-367. Bowles N, Poon L (1985) Aging and retrieval of words in semantic memory. Journal of Gerontology. 40, 1, 71-77. Bucks R, Singh S, Cuerden J et al (2000) Analysis of spontaneous, conversational speech in dementia of Alzheimer type: evaluation of an objective technique for analysing lexical performance. Aphasiology. 14, 1, 71-91. Carlomagno S, Santoro A, Menditti A et al (2005) Referential communication in Alzheimer s type dementia. Cortex. 41, 4, 520-534. Cherney L, Canter G (1993) Informational content in the discourse of patients with probable and patients with right brain damage. Clinical Aphasiology. 21, 123-134. Dijkstra K, Bourgeois M, Allen R et al (2004) Conversational coherence: discourse analysis of older adults with and without dementia. Journal of Neurolinguistics. 17, 4, 263-283. Drevenstedt J, Bellezza F (1993) Memory for self-generated narration in the elderly. Psychology and Aging. 8, 2, 187-196. Fleming V (2014) Early detection of cognitivelinguistic change associated with mild cognitive impairment. Communication Disorders Quarterly. 35, 3, 146-157. Fleming V, Harris J (2008) Complex discourse production in mild cognitive impairment: detecting subtle changes. Aphasiology. 22, 7-8, 729-740. Folkard S (1975) Diurnal variation in logical reasoning. British Journal of Psychology. 66, 1, 1-8. Folstein M, Folstein S, McHugh P (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. Glosser G, Deser T (1991) Patterns of discourse production among neurological patients with fluent language disorders. Brain and Language. 40, 1, 67-88. Groves-Wright K, Neils-Strunjas J, Burnett R et al (2004) A comparison of verbal and written language in. Journal of Communication Disorders. 37, 2, 109-130. Guendouzi J, Müller N (2008) Approaches to Discourse in Dementia. Lawrence Erlbaum Associates, Mahwah NJ. Hasher L, Chung C, May C et al (2002) Age, time of testing, and proactive interference. Canadian Journal of Experimental Psychology. 56, 3, 200-207. Horne J, Ostberg O (1977) Individual differences in human circadian rhythms. Biological Psychology. 5, 3, 179-190. Intons-Peterson M, Rocchi P, West T et al (1998) Aging, optimal testing times, and negative priming. Journal of Experimental Psychology: Learning, Memory, and Cognition. 24, 2, 362-376. Kemper S, Thompson M, Marquis J (2001) Longitudinal change in language production: effects of aging and dementia on grammatical complexity and propositional content. Psychology and Aging. 16, 4, 600-614. May C (1999) Synchrony effects in cognition: the costs and a benefit. Psychonomic Bulletin and Review. 6, 1, 142-147. May C, Hasher L, Stoltzfus E (1993) Optimal time of day and the magnitude of age differences in memory. Psychological Science. 4, 5, 326-330. May C, Hasher L, Foong N (2005) Implicit memory, age, and time of day: paradoxical priming effects. Psychological Science. 16, 2, 96-100. Miller J, Chapman R (2000) Systematic Analysis of Language Transcripts. University of Wisconsin, Madison WI. Morton L, Diubaldo D (1995) Circadian differences in hemisphere-linked spelling proficiencies. International Journal of Neuroscience. 81, 3-4, 101-110. Oakhill J (1986a) Effects of time of day and information importance on adults memory for a short story. The Quarterly Journal of Experimental Psychology Section A: Human Experimental Psychology. 38, 3, 419-430. Oakhill J (1986b) Effects of time of day on the integration of information in text. British Journal of Psychology. 77, 4, 481-488. Oakhill J (1988) Text memory and integration at different times of day. Applied Cognitive Psychology. 2, 3, 203-212. Orange J, Lubinski R, Higginbotham D (1996) Conversational repair by individuals with dementia of the Alzheimer s type. Journal of Speech and Hearing Research. 39, 4, 881-895. Reinberg A, Ugolini C, Motohashi Y et al (1988) Diurnal rhythms in performance tests of school children with and without language disorders. Chronobiology International. 5, 3, 291-299. Ripich D, Terrell B (1988) Patterns of discourse cohesion and coherence in. Journal of Speech and Hearing Disorders. 53, 1, 8-15. Ripich D, Carpenter B, Ziol E (2000) Conversational cohesion patterns in men and women with : a longitudinal study. International Journal of Language and Communication Disorders. 35, 1, 49-64. Santo Pietro M, Berman R (1984) Analysis of Connected Speech in Institutionalized Elderly with and without Senile Dementia. Paper presented at the annual conference of the American Speech Language and Hearing Association, San Francisco CA. Schmidt C, Collette F, Cajochen C et al (2007) A time to think: circadian rhythms in human cognition. Cognitive Neuropsychology. 24, 7, 755-789. Social Psychology Network (1997-2015) Research Randomizer. www.randomizer.org (Last accessed: March 6 2015.) Tomoeda C, Bayles K, Trosset M et al (1996) Cross-sectional analysis of Alzheimer disease effects on oral discourse in a picture description task. Alzheimer Disease and Associated Disorders. 10, 4, 204-215. Ulatowska H, Chapman S (1995) Discourse studies. In Lubinski R (Ed) Dementia and Communication. Singular Publishing, San Diego CA. Ulatowska H, Allard L, Donnell A et al (1988) Use of discourse analysis for evaluation of the condition of patients with dementia of the Alzheimer type. Polish Journal of Neurology and Neuroscience. 22, 1, 34-37. Winocur G, Hasher L (1999) Aging and time-of-day effects on cognition in rats. Behavioral Neuroscience. 113, 5, 991-997. Yaretsky A, Arzi T, Ashkenazi I (1995) Diurnal variation in language tests and its relevance to early detection of dementia: time dependence of language test in healthy adults. Journal of the American Geriatrics Society. 43, 2, 199-200. Yaretsky A, Arzi T, Ashkenazi I (1996) Diurnal variation in language tests and relevance to early detection of dementia: disruption of time dependency in dementing subjects. Journal of the American Geriatrics Society. 44, 7, 882-883. Yoon C, May C, Hasher L (1998) Aging, circadian arousal patterns, and cognition. In Schwarz N, Park D, Knäuper B et al (Eds) Cognition, Aging, and Self-Reports. Psychology Press, Philadelphia PA. 38 April 2015 Volume 27 Number 3 NURSING OLDER PEOPLE