Synthesis Essay Assignment

Change of Visual Perceptions in Taiwan’s Aged People with Mild Dementia Caused by the Textures of Building Materials

Yao-Rong Hwang

Published online: 1 February 2014 # Springer Science+Business Media New York 2014

Abstract People with dementia can be overstimulated by too many patterns and designs in one space. The purpose of this research was to find the visual perceptions’ changing of the demented elderly related to the textures of build- ing materials, and to inspect what kind of texture might have the possibility of causing demented elders to have visual hallucinations. A total of ten male subjects with mild dementia participated in this experiment. The simulation of visual perception was made using a highly sensitive LCD projector that showed pictures of building materials on a screen. Clock Drawing Test (CDT) was applied to assess the visual perceptions’ changing of the subjects before and after simulation. Based on the results of this experiment, the visual perceptions of the subjects were more changed by character textures and textures of regular shapes than by the other typologies of textures. Some of the subjects have the possibility of visual hallucinations while looking at the textures during the experiment, because they described visual images that did not exist. Data about these building materials can be made available for the reference of building managers and designers, in order to prevent the demented elderly from having behavior problems.

Keywords Dementia . Aged people . Visual perception . Simulation . Texture . Visual hallucination

Introduction

It is important to minimize the amount overstimulation in their surrounding environ- ments in order to prevent people with dementia from having hallucinations. Halluci- nations make demented individuals see and hear objects or images that do not exist

Ageing Int (2014) 39:221–232 DOI 10.1007/s12126-014-9196-9

Y.<R. Hwang (*) Department of Architecture & Interior Design, National Yunlin University of Science &Technology, 123, University Road, Section 3, Touliu, Yunlin, Taiwan, Republic of China 640 e-mail: [email protected]

Y. R. Hwang (*)

(Brawley 1997; Spencer et al. 2007) and may increase their level of distraction, agitation and confusion (Day et al. 2000; Alzheimer’s Association 2009). Visual hallucinations are caused by complex images that interfere with the output order of the sensory transmission (Pelaez 2000). Especially, for people with defect of cognition function, environmental messages that are improperly input to the sensory neural network may cause visual hallucinations.

Pattern and texture are components of every interior environment, with imagination there are many opportunities to introduce pattern and texture into the environment. Pattern can be difficult to use effectively, but it can also add a wealth of interest; however, we do know that people with dementia can be overstimulated by too many patterns and designs in one space (Brawley 1997). Furthermore, a pilot study reported that building materials with directional textures (e.g. textures with lines of different widths or ranges) resulted in the demented elderly changing their visual perceptions and seeing images that did not exist, which brought an extreme burden for caregivers (Hwang 2011).

Design guides call for appropriate levels of visual stimulation to prevent the demented elderly from having visual hallucinations, especially for mild demen- tia and moderate dementia, recommendations include removing unnecessary clutter, reducing glare, increasing color contrast and minimizing confusion arising from depth perception (Evans 1989; Brawley 1997). Although complex texture has been addressed as the issue in terms of visual hallucinations, little research on the impact of textures in dementia care environments was identified (Day et al. 2000; Bakker 2003).

It was found that the prevalence of hallucinations in people with dementia was 34.5 %, and that of these hallucinations 21.3 % were visual hallucinations in Taiwan (Lu and Lu 2000). The purpose of this research was to find the visual percep- tions’ changing of the demented elderly related to the textures of building materials except for directional textures, in order to inspect what kind of texture might have the possibility of causing demented elders to have visual halluci- nations, as well as to provide data about building materials as a reference for building managers and designers.

Methods

Participants

To perform the tests on the patients was required to have permission from the senior care centers and from the patients’ families, this made the search for research subjects become difficult. As there were fewer individuals with mod- erate and severe dementia residing in the caring facilities than there were individuals with mild dementia. A total of ten male subjects with mild demen- tia, represented as A, B, C, D, E, F, G, H, I and J, from the Chai-Li senior care center in Tainan City participated in this experiment. Permission to

222 Ageing Int (2014) 39:221–232

participate was given by the subjects’ families and the senior care center. The Clinical Dementia Rating Scale (CDR) and the Neuro-Psychiatric Inventory (NPI) were applied by neuro- psychiatric doctors to diagnosis the subjects’ degree of dementia and visual hallucinations. In addition, all of the subjects had normal vision, except for visual hallucinations.

Facility and Space of Visual Simulation

The simulation of visual perception was made using a highly sensitive LCD projector that showed pictures of building materials on a 300 cm wide and 280 cm high screen. The space for the experiment was located indoors and used artificial lighting and curtains to control the illumination. In order to match the simulation on the wall to the real item as much as possible, the distance between the screen and the projector was set at 478 cm, after several tests by the participants. The viewing distance was set at 300 cm from the viewing point to the screen after adjusting the field of view, so that the participants could see the whole picture and feel comfortable. The plan and evaluation of the projection and viewing distance are shown in Figs. 1 and 2. This experiment only processed the simulation on the wall due to the difficulty in projecting picture on the floor.

projector

view point

Fig. 1 Plan of picture projection and view distance

Fig. 2 Evaluation of picture projection and view distance

Ageing Int (2014) 39:221–232 223

Building Material Textures

It has been recognized that directional textures, periodic textures and random textures are obviously different (Efros and Leung 1999), and of these, random textures can be composed of or be free of picture images (Martin 2005; Ersboll and Pedersen 2007). Texture topology is classified into directional textures, periodic textures, random textures with picture images and random textures without picture images, whereas this research excluded directional textures due to the findings being addressed. Pattern recognition of the images was analyzed using a computer, and it was basically composed of the shape and distribution of the image elements (such as lines, blocks, and figures, etc.). The influencing factors on the shape were the size and direction of the image elements (big, small, vertical, horizontal, slope, etc.). As to the distribution of the image elements, range and density were the major dimensions for identifying the image patterns (Ersboll and Pedersen 2007; Stork 2009). There are a variety of classifications for the texture patterns of surface materials in buildings that are affected by the shape and distribution of elements (Martin 2005).

This research collected catalogues of material, including brick, stone, concrete, wood, tile, plastic, paint, clay, paper and carpet, from Taiwan building material association. A texture pattern analysis on the shape and distribution of the image elements, including the size, direction, range and density of the images, was processed after the collection of the material samples. According to the types of local building materials, the periodic texture was classified into regular lines and regular shapes (see Figs. 3 and 4). There were two categories of random textures with picture images, including flower figures and realistic figures (see Figs. 5 and 6). Finally, there were three categories for random textures without picture images, including irregular nature, natural imitation and character (see Figs. 7, 8 and 9). 3D Max and Photoshop7.0 software were applied to edit and scan the pictures from the catalogues, in order to produce texture images that were close to the real pictures.

Category

Regular vertical ( or horizontal) distribution

Regular slope distribution Regular vertical and horizontal crossed

distribution Fig. 3 Textures of regular lines (periodic textures)

Category

Same shape and density Different shape and density

Fig. 4 Textures of regular shapes (periodic textures)

224 Ageing Int (2014) 39:221–232

Category

Same density of flower distribution

Different density of flower distribution

Multi-flower assembly

Fig. 5 Textures of flower figures (random textures with picture images)

Category

Animal picture Plant picture Building picture Fig. 6 Textures of realistic figures (random textures with picture images)

Category

Natural stone Natural marble Fig. 7 Textures with an irregular nature (random textures without picture images)

Category

Wood imitation Brick imitation Marble imitation Stone imitation

Fig. 8 Textures that imitate natural textures (random textures without picture images)

Category

Different sizes of characters

Different types of characters

Different directions of characters

Fig. 9 Character textures (random textures without picture images)

Ageing Int (2014) 39:221–232 225

Clock Drawing Test (CDT)

For identifying and assessing visual hallucinations in older people with eye disease or cognitive impairment, Institute of Psychiatry Visual Hallucinations Interview (Santhouse et al. 2000) and North-East Visual Hallucinations Interview (Mosimann et al. 2008) were developed as screening tools. Except for participants only being questioned about their visual experience, Clock Drawing Test is a useful and widely supported screening tool that will indicate someone with good or poor visual-spatial ability and assist in the clinical diagnosis of hallucinations (Royall et al. 1998).

Clock drawing requires visual perception and visual movement to edit the positions of the numbers and the direction of the hands on a clock. As clock drawing requires several skills, including visual-spatial ability and constructional praxis, it can be used to investigate the visual space construction ability of demented elders (Samton et al. 2005; Peters and Pinto 2008). It is particularly sensitive to detect cognitive change, including visual perception, comprehension, memory, and executive functioning, among the elderly with neurological and psychiatric disorders (Shulman 2000). If the demented elderly have normal visual perceptions, they will draw the correct position of the numbers and the correct direction of the hands (Chiu and Tang 2006). Owing to most of the demented elderly having verbal problem, Clock Drawing Test has been applied to assess changes of visual-spatial ability and the unreal images seen by demented patients, in order to evaluate the possibility of visual hallucinations (Chuang 2008).

In general, during the clock-drawing test a piece of paper containing a circle is presented to the patient, on which he draw the numbers and the hands of a clock. The score of the test should be defined as every correct position of a number getting one point and the correct direction of the hands getting one point as well (Manos and Wu 1994). The hands should be clearly identified as being either the hour hand or the minute hand. In addition, numbers 6 and 12 must be on the vertical axis and numbers 3 and 9 must be on the horizontal axis. The positions of the other numbers will decide the rest of the scores. The total score is ten points, including eight positions for the numbers and two directions for the hands (see Fig. 10) (Chuang 2008). As to the direction of the hands, 11:10, 3:00 and 2:45 were applied in turn, to prevent the demented patients from becoming accustomed to the same draw- ing method.

A: 0 point B: 1 point – the position of 1 is correct. C: 2 points – the positions of 1 and 11 are correct. D: 4 points – the positions of 7, 8, 10 and 11 are correct. E: 8 points – the positions of 1, 2, 4, 5, 7, 8, 10 and 11 are correct, but it is

difficult to identify the hands. F: 10 points – the positions of 1, 2, 4, 5, 7, 8, 10 and 11 are correct, and the

direction of the hour hand and the minute hand are correct.

Fig. 10 Manos’ ten point score for the clock drawing test

226 Ageing Int (2014) 39:221–232

Simulation Procedure

During the visual perception simulation, the subjects drew clock before the simulation. Next, they looked at pictures of building materials and then drew the clock again. Researchers should present three to four pictures and a picture be projected on the wall for 20 s will be suitable for the simulation of an image (Chang 2006). This study presented one category each time, such as the category of regular shapes (see Fig. 4), which included four pictures containing same shape and density, and four pictures containing different shape and density. It took nearly 25 min for each subject to watch the sequence of eight pictures and to draw the clock. The subjects were interviewed by a caregiver, who asked about what kind of special images they had seen during the projection. Instructions and several practice of clock drawings were required to allow each subject to have a fine drawing ability before the formal test. It may be good evidence of visual hallucinations if the subjects draw what they have seen and that is unrelated to what was projected; however, this research was only able to find evidence through the interviews of the subjects owing to most subjects lacking any kind of drawing ability.

The experiment lasted for 21 weeks. A total of ten subjects participated in an experiment for each texture category. Experiments were held once every 3 weeks, in order to avoid the influence of visual perceptions that had changed during the previous experiment and in consideration of the learning effects of clock drawing in consecutive experiments. Experiments were divided into seven phases: periodic textures (two categories containing regular lines, and regular shapes), random textures with picture images (two categories containing flower figures, and realistic figures), and random textures without picture images (three categories containing textures with an irregular nature, textures that imitate natural textures, and character textures).

Description and Analysis

Descriptive and analytic statistics were performed using the SPSS computer program. A one-way analysis of variance was applied to compare the differences of clock drawing before simulation and after simulation, and significance was established using the paired- sample t test criterion. Another analysis was applied to compare the differences of clock drawing between different topologies and texture categories by using the independent samples t test criterion.

Results

Participant Demographics

With regards to the participants’ demographics, the ages of the subjects ranged from 75– 79 years old (50.0 %) to 80–85 years old (50.0 %). All of the participants were men. Regarding the subjects’ education level, 30 % had a university education and another

Ageing Int (2014) 39:221–232 227

30 % had a senior high school education, followed by junior high school (20.0 %) and elementary school (20.0 %). All of the subjects suffered from mild demen- tia, and 50.0 % had psychiatric symptoms, while the other 50.0 % did not have psychiatric symptoms.

Differences of Clock Drawing

One of the subjects, whose visual perceptions had been influenced by the simulation, was selected from each typology of texture to describe the differences of the clock drawing (see Table 1). The scores of the clock drawing test (see Table 2) showed that the visual perceptions of the subjects were more changed by character textures (eight subjects were influenced) and textures of regular shapes (six subjects were influenced) than by the other textures. On the other hand, the subjects were less changed by textures of realistic figures (zero subject was influenced). With regards to individual cases, cases A, E and F were more changed by simulation than were the other cases, except for random textures with realistic figures. On the other hand, cases C, G, H and J were less

Table 1 The differences of clock drawing before and after the simulation of visual perceptions

Typology of Texture Case Before

simulation After

simulation Difference of Clock Drawing

Periodic textures (textures of regular

lines)

Case A

clock drawing: 2: 45 before : The position of 5 was incorrect, and the minute hand was shorter than the hour hand. after : The positions of 8, 10 and 11 were incorrect, as was the direction of the minute hand.

Periodic textures (textures of regular

shapes)

Case F

clock drawing: 3: 00 before : The positions of 8, 10 and 11 were incorrect, as was the direction of the minute hand. after : The positions of all numbers were incorrect, and it was difficult to identify the hands.

Random textures with picture images

(textures of flower figures)

Case A

clock drawing: 2: 45 before : The position of 5 was incorrect. after : The positions of 4 and 5 were incorrect.

Random textures with picture images

(textures with an irregular nature)

Case E

clock drawing: 11: 10 before : Correct drawing. after : The positions of the hands were incorrect.

Random textures without picture images (textures that imitate

natural textures)

Case I

clock drawing: 2: 45 before : Correct drawing. after : The direction of the minute hand was incorrect.

Random textures without picture images

(character textures)

Case D

clock drawing: 11: 10 before : Correct drawing. after : Position 1 was missing, and it was difficult to identify the hour hand.

228 Ageing Int (2014) 39:221–232

changed by simulation than were the other cases, except for random character textures without picture images (see Table 2).

Discussion

Table 3 shows that the average score was 7.30 before the simulation of visual perceptions and was 6.20 after the simulation of visual perceptions when the subjects looked at periodic textures with regular shapes. The difference in the average score between pre-simulation and post-simulation was significant (t=1.859, p=0.048<0.05). Furthermore, the difference in the average score between pre-simulation and post-

Table 3 Average score of clock drawing before and after the simulation of visual perceptions using Paired- Sample t Test

Score Number of samples

Average score

Variation of score

t p

Typology of texture

Periodic textures with regular shapes (before) 10 7.30 2.335 1.859 0.048*

Periodic textures with regular shapes (after) 10 6.20 1.911

Random character textures without picture images (before)

10 8.60 2.025 2.011 0.039*

Random character textures without picture images (after)

10 7.40 1.443

p <0.05* p <0.01** p <0.001***

Table 2 Score statistics of the clock-drawing test

Case A B C D E F G H I J Subjects influenced

Typology of texture

Periodic textures Regular lines Before 8 7 6 9 8 7 9 9 9 7 3

After 6 7 7 9 8 6 9 8 9 7 Subjects

Regular shapes Before 8 10 4 6 10 6 7 6 8 8 6

After 6 9 5 5 9 0 7 6 7 8 Subjects

Random textures with picture images

Flower figures Before 10 9 8 9 8 7 7 8 9 9 3

After 9 9 8 9 7 7 7 8 9 8 Subjects

Realistic figures

Before 7 6 7 6 9 8 8 7 8 7 0

After 7 6 7 7 9 8 8 8 8 7 Subjects

Random textures without picture images

Irregular nature Before 9 7 7 10 10 10 9 6 9 8 3

After 9 7 7 10 8 9 9 6 8 8 Subjects

Imitating natural textures

Before 9 8 10 9 9 7 7 7 10 6 4

After 8 8 10 9 8 6 8 7 9 6 Subjects

Character textures

Before 9 10 10 10 9 8 9 7 8 6 8

After 7 8 9 8 7 7 8 7 7 6 Subjects

Ageing Int (2014) 39:221–232 229

simulation was significant (t=2.011, p=0.039<0.05) when the subjects looked at random character textures without picture images.

The differences in the average scores of clock drawing after the simulation of visual perceptions among various typologies of textures showed no significance (see Table 4), even the average scores of clock drawing in terms of periodic textures were obviously less than those of random textures without picture images. On the other hand, the average scores before and after the subjects looked at textures from the first phase to seventh phase did not show that their clock drawings were getting better (see Table 2). This mean that the learning effects of clock drawing did not happen in this consecutive experiments, and the scores of clock drawing really showed the possibility of visual perceptions’ changing.

The differences of the clock drawings showed that the visual perceptions of the subjects were more changed by character textures (random textures without picture images) and textures of regular shapes (periodic textures) than by the other textures, either from the number of influenced subjects or from the comparison of the average scores. It was found that some of the subjects might have experienced visual halluci- nations while they looked at the textures in this experiment, because they described visual images that did not exist during the interview with the caregiver. Regarding character textures, case A had a vision of “many warms moving and I felt uncomfort- able”, and case E was excited about having a vision of “many fruit of different shapes hanging over there”. Case F had a vision of “many blocks distributed on the wall… different sizes” while he looked at textures of regular shapes. The phenomenon of visual perceptions’ changing caused the subjects to see visual images that did not exist, which was similar to the findings in the pilot study that showed the influence of directional textures on the change of visual perceptions (Hwang 2011).

Due to the difficulty in getting permission to perform the experiment, as well as the Chai-Li senior care center only containing male residents, this study was unable to investigate the variable of gender. The age of the patient and the severity of disease have always been recognized as variables of the behavioral problems in the demented elderly. All of the participants in this experiment were over 75 years old and the severity of their dementia was mild. There were limitations on comparing younger

Table 4 Average score of clock drawing after the simulation of visual perceptions among typologies of textures using the Independent-Sample t Test

Score Number of samples

Average score

Variation of score

t p

Typology of texture

Periodic textures 20 6.90 2.175 −1.172 0.234 Random textures with picture images 20 7.80 1.303

Periodic textures 20 6.90 2.175 −1.087 0.271 Random textures without picture

images 30 7.83 1.367

Random textures with picture images 20 7.80 1.303 −0.044 0.960 Random textures without picture

images 30 7.83 1.367

p <0.05* p <0.01** p <0.001***

230 Ageing Int (2014) 39:221–232

elderly with older elderly and on discussing the differences between mild dementia and moderate dementia in this experiment. Literature demonstrated the strong correlation of visual dysfunction with the severity of disease, as well as its relatively weaker correlation with age in patient groups (Tabassi et al. 2003). Nevertheless, it was found that the average age of demented elders who had visual hallucinations was older than that of demented elders without visual hallucination (Soeda et al. 2004). Obviously, further study is required to confirm if the influence on the visual hallucinations of the demented elderly correlates to the variables of age, gender and the severity of disease.

Conclusion

Based on the results of this experiment, the four aspects of the conclusion were as follows. First, the visual perceptions of the subjects were more changed by character textures and textures of regular shapes than by the other typologies of textures. Second, some of the participants have the possibility of visual hallucinations while looking at the textures during the experiment, because they described visual images that did not exist. Next, data about these building materials should be made available for the reference of building managers and designers, in order to prevent the demented elderly from having behavior problems due to environmental design. Lastly, there were limitations on comparing the younger elderly with the older elderly and on discussing the differences between mild dementia and moderate dementia in this experiment. Further study is required to confirm if the influence on visual perceptions in the demented elderly correlates to the variables of age, gender and the severity of disease.

Acknowledgments This research was made possible thanks to budgetary support from National Yunlin University of Science & Technology and Chai-Li senior care center that offered demented individuals for conducting this experiment.

References

Alzheimer’s Association. (2009). Dementia care practice: Recommendations for professionals working in a home setting. Chicago: Alzheimer’s Association.

Bakker, R. (2003). Sensory loss, dementia, and environments. Journal of Generations, 27(1), 46–51. Brawley, E. C. (1997). Designing for Alzheimer’s disease : Strategies for creating better care environments.

New York: John Wiley & Sons Inc. Chang, S. S. (2006). Exploring passenger cars’ form identity through visual images and perception. Tainan:

Unpublished Thesis. Cheng Kung University Press. Chiu, M. J., & Tang, L. Y. (2006). The guideline of dementia care. Taipei: H2O Book Publishing

Company. Chuang, M. F. (2008). Clock drawing test as a feasible tool for identifying behavioral and psychological

symptoms of dementia. Hualien: Tsz-ji University Press. Day, K., Carreon, D., & Stump, C. (2000). The therapeutic design of environments for people with dementia: a

review of the empirical research. The Gerontologist, 40(4), 397–416. Efros, A. A., & Leung, T. K. (1999). Texture synthesis by non-parametric sampling (pp. 1033–1038). Corfu:

IEEE International Conference on Computer Vision.

Ageing Int (2014) 39:221–232 231

Ersboll, B. K., & Pedersen, K. S. (2007). Lecture notes in computer science: Image processing, computer vision, pattern recognition and graphics. New York: Springer Scientific Publishing Service.

Evans, B. (1989). Managing from day to day: Creating a safe and workable environment. Minneapolis: Department of Veterans Affairs Medical Center.

Hwang, Y. R. (2011). The possibility of visual hallucinations in the demented elderly caused by the textures of building materials –-A pilot study of building materials with directional textures. Yulin: National Yunlin University of Science &Technology Press.

Lu, Y. C., & Lu, H. G. (2000). Cognitive symptoms of Alzheimer’s disease. Research of Applied Psychology, 7, 104–116.

Manos, P. J., & Wu, R. (1994). The ten point clock test: a quick screen and grading method for cognitive impairment in medical and surgical patients. International psychiatry Medicine, 24, 229–244.

Martin, C. (2005). The surface texture book. London: Thames & Hudson Ltd. Pelaez, J. R. (2000). Towards a neural network based therapy for hallucinatory disorders. Neural Network, 13,

1047–1061. Mosimann, U. P., Collerton, D., Dudley, R., Meyer, T. D., Graham, G., & Dean, J. L. (2008). A semi-

structured interview to assess visual hallucinations in older people. International Journal of Geriatric Psychiatry, 23(7), 712–718.

Peters, R., & Pinto, E. M. (2008). Predictive value of the clock drawing test. Dementia geriatric Cognitive Disorder, 26, 351–355.

Royall, D. R., Cordes, J. A., & Polk, M. (1998). CLOX: an executive clock drawing task. Journal of Neurology, Neurosurgery & Psychiatry, 64(5), 588–594.

Samton, J. B., Ferrando, S. J., Sanelli, P., Karimi, S., Raiteri, V., & Barnhill, J. W. (2005). The clock drawing test: diagnostic, functional, and neuroimaging correlates in older medically ill adults. The Journal of Neuropsychiatry & Clinical Neurosciences, 17(4), 533–540.

Santhouse, A. M., Howard, R. J., & Fytche, D. H. (2000). Visual hallucinatory syndromes and the anatomy of the brain. Brain, 123(10), 2055–2064.

Shulman, K. I. (2000). Clock-drawing: is it the ideal cognitive screening test? International Journal of Geriatric Psychiatry, 15(6), 548–561.

Soeda, S., Terao, T., Nishimura, M., & Nakamura, J. (2004). Aging and visual hallucinations in elderly psychiatric outpatients. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 28(2), 401– 404.

Spencer, B., Robinson, A., & Curtin, C. (2007). Developing meaningful connections with people with dementia: A training manual. Ann Arbor: University of Michigan Press.

Stork, D. G. (2009). Computer vision and computer graphics analysis of paintings and drawings: An introduction to the literature. New York: Springer Scientific Publishing Service.

Tabassi, S. A., Sepanlou, S. G., & Lotfi, J. (2003). Acquired color vision deficiency in Parkinson ’s disease. Acta Medica Iranica, 41(3), 143–146.

Yao-Rong Hwang is an associate professor in the Department of Architecture and Interior Design at the National Yunlin University of Science & Technology, Taiwan, ROC. He studied the environmental design for the aged people since 1990, and article entitled “Housing for the Elderly in Taiwan” was published ( Aging International, 1997, Volume XX III, 133–147). Hwang recently focused on the designing issues of special care unit for the demented elderly and had five oral presentations in terms of dementia care environment in the EDRA ( Environmental Design Research Association ) international conference. Paper presentation has ever been selected for full paper publication in the proceeding.

232 Ageing Int (2014) 39:221–232

Copyright of Ageing International is the property of Springer Science & Business Media B.V. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.