- Email: [email protected]
Programmed Instruction to teach pointing with a computer mouse in preschoolers with developmental disabilities
Research in Developmental Disabilities 27 (2006) 175–189
Programmed Instruction to teach pointing with a computer mouse in preschoolers with developmental disabilities Hirofumi Shimizu a,b,*, Christopher S. McDonough b a
b
Meisei University, Tokyo, Japan Hawthorne Country Day School, 5 Bradhurst Avenue, Hawthorne, NY 10532, USA
Received 30 September 2004; received in revised form 20 January 2005; accepted 23 January 2005
Abstract Programmed Instruction combined with experimenter-provided prompts (physical, verbal, and gesturing) was used to teach pointing with a computer mouse. Three preschoolers who scored at least 1 year below their chronological age levels participated. During the pre-assessment, none of the participants demonstrated pointing. However, they could press and release the mouse button. Programmed Instruction consisted of three stages, based on an analysis of the behavioral prerequisites for pointing. Stage 1 was designed to teach participants to move the mouse. Stage 2 was designed to teach participants to move the on-screen cursor onto specific items on the screen. Stage 3 was designed to teach participants to click on specific items on the screen. Experimenter-provided prompts were used to facilitate skill acquisition at each stage. The post-assessment showed that all participants learned pointing after intervention. The intervention package consisting of Programmed Instruction and experimenter-provided prompts was effective for teaching the hand–eye coordination required for pointing. # 2005 Elsevier Ltd. All rights reserved. Keywords: Computer-based instruction; Developmental disabilities; Mouse; Programmed Instruction; Preschooler
* Corresponding author. Tel.: +1 914 592 8526x166; fax: +1 914 592 3227. E-mail address: [email protected] (H. Shimizu). 0891-4222/$ – see front matter # 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ridd.2005.01.001
176
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
Computer-based instruction is widely used in special education (e.g., Lancioni & Boelens, 1996; Leung, 1994; Mastropieri, Scruggs, & Shiah, 1997; Merzenich et al., 1996; Tallal et al., 1996), including the field of Applied Behavior Analysis (e.g., Dube, Iennaco, Rocco, Kledaras, & McIlvane, 1992; Dube, McDonald, & McIlvane, 1991; Lane & Critchfield, 1998; Neef, Bicard, & Endo, 2001; Saunders, Johnston, Tompkins, Dutcher, & Williams, 1997; Stromer, Mackay, Howell, & McVay, 1996). In the previous reports, many researchers have used a touch panel as a computer-input device. Although there might be other reasons for researchers to use a touch panel (e.g., automated data collection for selection-based response classes), special education teachers may sometimes encounter difficulty teaching their students to use a mouse. For such students, unless a computer is equipped with a specialized alternative input device such as a touch panel, their access to the growing number of well-designed educational programs available to computer users is limited. A mouse is one of the most familiar input devices. This device can be used for almost all computers. The movement of an arrow-shaped cursor on the computer screen is achieved by moving the mouse. People can interact with a computer by pressing and releasing the button of the mouse. The two basic mouse actions most commonly used are pointing and dragging. For pointing, people move the mouse in order to move a cursor on the computer screen. When the cursor is pointed at a specific item such as an icon, picture, or text, they press and release the mouse button to send functional commands to the computer. Pressing and releasing the mouse button to send functional commands is often called click or clicking. In dragging, the mouse button is pressed and held down when the cursor is on a specific item on the screen and then the mouse is moved while the button remains held down. When the cursor reaches another specific item the mouse button is released sending a command to the computer. Dragging is often used to draw, move, or highlight objects. Previous studies have shown that normal four-year-old preschoolers demonstrate pointing and dragging (e.g., King, 1992). Even three-year-old children show the ability to use a mouse (e.g., Stronmen, Revelle, Medoff, & Razavi, 1996). Pointing is easier than dragging for both children (e.g., Joiner, Messer, Light, & Littleton, 1998) and adults (e.g., Mackenzie, 1992). Some studies have indicated that children with developmental disabilities have the ability to learn pointing (e.g., Durfee & Billingsley, 1998; Missiuna, 1994; Shimizu & Yamamoto, 2000). For example, Durfee and Billingsley (1998) reported that a mouse was a more effective interface device than a touch panel for a boy with spastic quadriplegic cerebral palsy. Shimizu and Yamamoto (2000) assessed the pointing skill of 16 students with severe and moderate mental retardation ranging in age from 7 to 12 years old. Eight students demonstrated pointing without any problems, four students demonstrated gradual improvement throughout the assessment and four students could not use a mouse. In addition to these findings, we have found in our laboratory that many preschoolers who are identified as ‘‘preschoolers with disabilities’’ demonstrate the ability to use a mouse. However, some do not. The number of children who are able to use a mouse might increase if the teaching process is highly controlled with the application of behavioral technology. The purpose of the present study was to develop and evaluate an instructional sequence (intervention) to teach pointing with a computer mouse to three preschoolers with developmental disabilities. The strategy had three characteristics: (a) breaking the terminal
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
177
skill into several component skills, (b) shaping each component and systematically integrating them, and (c) errorless teaching throughout instruction. The intervention consisted of Programmed Instruction and experimenter-provided prompts (physical, verbal, and gesturing). The terminal skill, pointing, was broken down into four component skills: (a) pressing and releasing the mouse button, (b) moving the mouse, (c) pointing the cursor at specific items on the screen, and (d) clicking on specific items on the screen. Since all of the participants demonstrated the first component skill (pressing and releasing the mouse button) during pre-assessment, instruction was not provided for this component skill. The instruction consisted of three stages to shape each of the other component skills and to systematically integrate them. The first stage of training was designed to shape movement of the mouse. The second stage was designed to teach pointing the cursor at specific items on the screen, a skill requiring precise hand–eye coordination. Because of the complexity of this skill it was broken down into four sub-components: (a) looking at the cursor, (b) looking at specific items on the screen where the cursor was to be moved, (c) moving the mouse to point the cursor at specific items, and (d) keeping the mouse in the correct position (i.e., facing the computer). The third stage was designed to teach participants to click on the specific items on the screen. To promote errorless learning, Programmed Instruction was supported by experimenter-provided prompts.
1. Method 1.1. Participants Three 4-year-old children with developmental disabilities participated: Beth, Mike and Taylor. Each participant was identified as a ‘‘preschooler with a disability’’ by his or her school district’s Committee on Preschool Special Education. The children were enrolled in the special education preschool program where this study was conducted. All participants scored at least 1 year below their chronological age level on standardized assessments. Their dominant arms were right. These children could press and release the mouse button before the training, but could not use the mouse for pointing. 1.2. Apparatus and setting Instruction occurred in brief sessions (average 14 min) in a quiet room at the preschool one to four times per week. The computer, an Apple iMac1, presented stimuli and recorded data while the experimenters provided and recorded prompts. The computer software for assessment was developed with Apple Hypercard1 2.2, while the software for training was created with Macromedia Director1 8.0. The computer input device was an Apple USB Mouse. The terminal goal was for participants to acquire pointing using this mouse. Participant Mike used two other mice in the process (Stage 2-2). One mouse was a Macally imouse1, and the other was a Logitech Cordless Wheel. Taylor also used a Macally imouse1 during Stage 2-2. In addition to the mice, Mike used a touch panel (Troll Touch TouchStar) at the beginning of Stage 2-2.
178
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
The computer was placed on a 1 m 1 m table. The participants sat on a chair, facing the computer screen. There was space (30 cm 60 cm) on the table between the participant and the computer, and participants manipulated the mouse in this area. The keyboard was located on the left side of the computer out of the participant’s reach. The two experimenters sat behind and to either side of the participant. 1.3. Assessment for pointing 1.3.1. Procedure Assessment occurred before and after training. The assessment tool used was the computer software reported by Shimizu and Yamamoto (2000). Fig. 1 illustrates the computer screens of the assessment. Assessment began with a click on the ‘‘START’’ button by the experimenter (Fig. 1A). The computer presented the auditory instruction: ‘‘What’s hiding behind the squares?’’ The screen was filled with various sizes of 15 black rectangles (Fig. 1B) and a picture was ‘‘hiding’’ behind the rectangles. Although there were small gaps between the rectangles, the picture was not in view. It resembled a puzzle placed facedown. As the participants pointed the on-screen cursor at the rectangles and pressed the mouse button (clicked on the rectangles), the clicked rectangle disappeared with a brief chime, revealing part of the picture (Fig. 1C and D). After all of the rectangles had been clicked the whole picture appeared with a brief musical fanfare (Fig. 1E). The opportunity to click on one rectangle was defined as one trial. Clicking on the 15 rectangles was defined as one block. After one block was completed, the next block began with the presentation of 15 more black rectangles and the auditory instruction: ‘‘What will you find next?’’ Five consecutive blocks were conducted during the pre-assessment and
Fig. 1. The sample computer screens for the pre-assessment and the post-assessment phases. Panel A: The assessment began with the ‘‘START’’ button pressed. Panel B: At the beginning of one block, 15 black rectangles appeared on the screen. A picture was hidden behind the rectangles. Panels C and D: As participants clicked on the rectangles, the clicked rectangle disappeared, revealing part of the picture. Panel E: As they clicked on all the rectangles, the whole picture appeared. Panel F: The assessment ended.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
179
post-assessment. The picture behind the rectangles changed from block to block. After five blocks were completed, the word ‘‘END’’ appeared on the screen with a short jingle (Fig. 1F). The experimenters used a delayed prompting procedure (e.g., Touchette & Howard, 1984) during assessment. If the participants did not click on a rectangle after 5 s, one of the experimenters guided their performance with one of three kinds of prompts. (a) Physical prompt: An experimenter helped the participant move the mouse or press the mouse button (or both), holding the participant’s right hand or arm. (b) Verbal prompt: An experimenter said, ‘‘Move it’’ or ‘‘Push the button’’. (c) Gesturing to the screen: An experimenter pointed his index finger at the rectangles on the computer screen. 1.3.2. Measurement Four measures were used to assess pointing. (a) Total number of clicks per block: One click was defined as one press of the mouse button. (b) Total time per block: This measure was defined as the length of time it took for the participant to click on all 15 black rectangles. Time recording began when the rectangles were present on the screen and stopped after the last rectangle was clicked. (c) Total length of on-screen cursor movement per block: This measure was the combined length of all cursor movement during the block. The measurement unit was recorded in pixels. A pixel is a unit used to describe the resolution of a computer screen. One pixel is the smallest element. For example, 640 480 resolution means that the screen consists of the 640 pixels in width and 480 pixels in height. The resolution of the computer display in this experiment was 640 480. If a participant moved the cursor from one side of the screen to the other, the computer recorded the length as 640 pixels. All cursor movement during the block was recorded and summed to determine the total length of on-screen cursor movement. (d) Total number of trials per block in which the experimenters provided prompts was the fourth measure. 1.3.3. Interobserver agreement Interobserver agreement related to prompting was calculated between the two experimenters. They independently recorded the incidence of prompting during assessment sessions in real time. Agreement was calculated by dividing agreements by agreements plus disagreements and multiplying by 100%. Interobserver agreement was 96%. The computer automatically recorded other dependent measures. 1.4. Programmed instruction procedures Table 1 illustrates a training goal and an intervention package at each stage of the programmed instruction. The programmed instruction consisted of three stages and was designed to teach and integrate: (a) moving the mouse; (b) pointing the cursor at specific items on the screen; and (c) clicking on specific items on the screen. 1.5. Stage 1: moving the mouse Stage 1 was designed to teach moving a mouse. All three participants pressed and released the mouse button frequently during pre-assessment and the initial teaching of
180
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
Table 1 Training goal and experimenter-provided prompts Stage
Goal
Experimenter-provided prompts
1
Moving the mouse
1. Physical prompts to help move the mouse
2
Pointing the cursor at specific items on the screen
1. Physical prompts to help move the mouse 2. Verbal prompts (e.g., ‘‘move the mouse’’ and ‘‘look at this’’) 3. Gesturing to the screen: Pointing trainer’s index finger to the cursor or rectangles 4. Keeping the mouse in the correct position (this was used only in Stage 2-2)
3
Clicking on specific items on the screen
1. Physical prompt to help move the mouse and press the mouse button 2. Verbal prompts (e.g., ‘‘Press the button’’, ‘‘Push the button’’)
Stage 1. Since this appeared to interfere with shaping movement of the mouse, the mouse button was disabled so the participants could not press and release the button. A completely black screen was presented at the beginning of each trial (Fig. 2A). The screen consisted of 25 black rectangles, all the same size (128 96 pixels). A picture was hidden behind the rectangles. If the participants moved the mouse, three forms of feedback followed the movement (Fig. 2B): (a) one rectangle disappeared and the part of the picture corresponding to that rectangle was revealed, (b) the color of the remaining rectangles
Fig. 2. The sample computer screens for Programmed Instruction. Panel A (Stages 1 and 2): At the beginning of the trial, a completely black screen was presented. The screen consisted of 25 black rectangles, all the same size. A picture was hidden behind the rectangles. Panel B (Stage 1): As participants moved the mouse in Stage 1, the rectangles disappeared and the color of the remaining rectangles changed with a brief sound. Panels C and D (Stage 1): The more participants moved the mouse, the more rectangles disappeared, the more color changes occurred in the remaining rectangles, and the more the participants heard the brief sound. Panel E (Stage 2-2): The size of the cursor was enlarged. The participants were required to point the cursor at rectangles. Panel F (Stage 22): Whenever the cursor moved for 1 pixel on the screen, the color of the cursor randomly changed. When the cursor was pointed at rectangles they disappeared with a brief chime.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
181
changed, and (c) a brief sound was presented. The computer randomly determined which rectangle disappeared and the color of the remaining rectangles. The more often the participants moved the mouse, the more rectangles disappeared, the more color changes occurred in the remaining rectangles, and the more participants heard the brief sound (Fig. 2C and D). When all the rectangles disappeared, the entire picture was revealed with a brief musical fanfare. Clearing all 25 rectangles to reveal the picture was defined as one trial. The detection of the mouse movement by the computer depended on the length of the on-screen cursor movement measured in pixels. At the beginning of Stage 1, the detected length of pixels was set to 1 pixel. One pixel movement of the on-screen cursor represented approximately 0.6 mm movement of the mouse on the table. In other words, when the participant moved the mouse on the table for 0.6 mm, the cursor moved for 1 pixel on the screen, and one black rectangle disappeared. Therefore, at the beginning of Stage 1, one trial was defined as the clearing of all 25 rectangles, which was equivalent to moving the cursor on the screen for 25 pixels and the mouse for approximately 15 mm. The detected length increased pixel by pixel, depending on the participant’s progress. For example, if the participant met criteria at the trial with a detection length of 1 pixel, the experimenters set the detection length to 2 pixels for the next trial. The criteria to increase the pixel detection length were that (a) the participant moved the mouse without any prompts, and (b) the total time to finish the trial was less than 25 s. As illustrated in Table 1, the experimenter provided physical prompts at Stage 1. If the participant could not move the mouse, the experimenters prompted the participant to move the mouse, holding their right hand or arm. Whenever one trial finished without any prompts, the experimenters applauded the participant with words or phrases (e.g., ‘‘wonderful’’, ‘‘excellent’’, and ‘‘good job’’). Stage 1 ended when the participant met the criteria and they spontaneously moved the mouse on the table more than 30 cm. 1.6. Stage 2: pointing the cursor at specific items on the screen Stage 2 was designed to teach pointing the on-screen cursor at specific items on the screen. This skill requires relatively complex hand–eye coordination. It requires (a) looking at the cursor, (b) looking at specific items on the screen where the cursor is to be moved, (c) moving the mouse to point the cursor at specific items, and (d) keeping the mouse in the correct position, which is pointed toward the computer. Stage 2 was divided into three parts: 2-1, 2-2, and 2-3. Stage 2-2 was developed in response to two participants failing to meet criteria at Stage 2-1 (described below). The mouse button remained disabled throughout Stage 2 so that the participant could not press and release the button. 1.6.1. Stage 2-1 Stage 2-1 began like Stage 1 with the same black screen consisting of 25 black rectangles. The main difference between Stages 1 and 2-1 was that in Stage 2-1 the participants were required to point the cursor at each of the rectangles on the screen. In Stage 1, it did not matter if the cursor pointed to each rectangle or not, only that it moved a predefined number of pixels. In Stage 2-1, when the cursor was pointed at a rectangle, that rectangle disappeared with a brief chime and a portion of the picture appeared in its place.
182
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
The other difference between Stages 1 and 2-1 was that the color of the remaining rectangles did not change with cursor movement. The rectangles always remained black. In Stage 2-1 one trial was defined as moving the cursor so it pointed at each of the 25 rectangles on the screen. When necessary, the experimenters used prompts during Stage 2-1 as described in Table 1. If the participants did not move the mouse after 5 s, one of the experimenters prompted the participants to move the mouse either physically or verbally. Furthermore, if the participants did not look at the cursor or rectangles, one of the experimenters pointed their index finger to the cursor or rectangles and said, ‘‘Look at this’’. When all 25 rectangles disappeared or one trial was completed without any prompts, the experimenters applauded participants’ performance with words or phrases. The criteria for Stage 2-1 were that (a) the participants pointed the cursor at all rectangles without any prompts in three consecutive trials, and (b) the total time to finish those trials was less than 15 s. Only Beth could meet these criteria. 1.6.2. Stage 2-2 Stage 2-2 was developed because Taylor and Mike failed to meet Stage 2-1 criteria. During Stage 2-1 Taylor and Mike (especially Mike) were observed to have trouble controlling the mouse. That is, the mouse would become rotated to a sideways or upside down position. When the mouse is out of position relative to the computer screen it becomes difficult to operate because the cursor movement is not reflective of the mouse movement. In addition, Taylor and Mike required frequent gestural and verbal prompts to continue looking at the on-screen cursor. This could indicate that they did not consistently observe the on-screen cursor during Stage 2-1, an observation without which they had no chance at acquiring the target skill. Therefore, to assist Taylor and Mike with observation of the on-screen cursor and control of the mouse, Stage 2-2 was developed. In Stage 2-2 the initial size of the on-screen cursor was about 800% (Fig. 2E and F) of normal size (11 16 pixels, which is the size used throughout the experiment). In addition, the computer was programmed to randomly change the color of this enlarged cursor whenever it moved 1 pixel on the screen. The cursor size gradually decreased across 10 steps depending on the participant’s progress. The criteria to change the cursor size were that (1) the participant pointed the cursor at all rectangles without prompts, and (2) the total time to finish the trial was less than 20 s. Other Programmed Instruction features of Stage 2-2 were identical to Stage 2-1. Two things were done in an effort to help Taylor and Mike control the mouse. First, the mouse cord was shortened so that moving the mouse out of position became difficult. This was done only for Mike. The length of the mouse cord also was manually controlled by keeping it without slack while Mike was moving the mouse. Second, two different mice were used with Mike (a Macally imouse1, which has an oval shape unlike the round Apple USB Mouse, and a Logitech Cordless Wheel Mouse) and one with Taylor (the Macally imouse1). In addition, the prompts used in Stage 2-1 remained in effect. Mike was allowed to use a touch screen for three trials in Stage 2-2 in an effort to help him look at the on-screen cursor and to experience independent success with pointing to
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
183
rectangles on the computer screen. The hand–eye coordination required for pointing to items on the computer screen with one’s finger is not as difficult as using the mouse to control the on-screen cursor. When using the touch screen, the enlarged on-screen cursor moved to the point where Mike’s finger contacted the screen. 1.6.3. Stage 2-3 Stage 2-3 was the same as Stage 2-1 except that the screen used during 2-3 was the same as the screen used during assessment (Fig. 1B). If the participants pointed the cursor at the rectangles they disappeared, with a brief chime, and part of the picture was revealed. The criteria to complete Stage 2-3 were that (1) the participant pointed the cursor at all rectangles without any prompts in two consecutive trials, and (2) the total time to finish those trials was less than 15 s. 1.7. Stage 3: teaching clicking on specific items on the screen Stage 3 was designed to teach clicking on specific items on the computer screen. The mouse button was enabled at this stage. The computer program at Stage 3 was the same one used for assessment, however, the program only recorded duration. The participants were required to point the cursor at rectangles, and to click on them. The experimenters prompted them to point the cursor at rectangles, and click on rectangles, using physical prompts or verbal prompts (e.g., ‘‘Push the button’’, ‘‘Press the button’’). When rectangles disappeared or one trial was completed without any prompts, the experimenters praised the participant. The criteria to complete Stage 3 were that (1) the participants clicked on all 15 rectangles without any prompts in three consecutive trials, and (2) the total time to finish those trials was less than 30 s.
2. Results 2.1. Assessment Fig. 3 shows the results of the pre-assessment and the post-assessment for each participant across four dependent measures. Pre-assessment scores and post-assessment scores are shown as gray and white bars, respectively. Five blocks were conducted during pre-assessment and post-assessment and each bar represents the average of five blocks. The Y error bars (vertical hairlines stuck into the top of each bar) represent the range of the data in the five blocks. The top three panels of Fig. 3 show the results of the total number of clicks per block. Fifteen was the minimum number of clicks required to clear the rectangles and reveal the entire picture because there were 15 rectangles. All participants could press and release the mouse button at the pre-assessment. Taylor and Mike showed very high numbers of clicks at the pre-assessment (ranged from 47 to 61 for Taylor and from 94 to 138 for Mike) while the number of clicks for Beth ranged from 22 to 35. Performance greatly improved after intervention. The number of clicks at the post-assessment ranged from 15 to 16 for Beth, 18 to 21 for Taylor, and 19 to 22 for Mike.
184
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
Fig. 3. The results of the pre-assessment and the post-assessment phases. Pre-assessment scores and postassessment scores are shown as gray and white bars, respectively. Five blocks were conducted at both preassessment and post-assessment and each bar represents the average of five blocks. The Y error bars (vertical hairlines stuck into the top of each bars) represent the range of the data in the five blocks.
The second three panels are the result of the total time per block. All the participants took about 2 min to finish one block at the pre-assessment. The speed was drastically improved after intervention. They each took about 20 s to finish one block. The third three panels show the results of the total length of cursor movement per block. The data from a previous study indicate that the length of cursor movement is around 3800 pixels on average, if students with disabilities demonstrate appropriate pointing (Shimizu & Yamamoto, 2000). Beth and Mike scored 1416 pixels and 2807 pixels, respectively, on average at the pre-assessment, while Taylor scored 3591 pixels. Beth,
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
185
Table 2 The total number of trials to meet criteria at each stage, total number of sessions, and total time for intervention Participants
Beth Taylor Mike
The total number of trials Stage 1
Stage 2
Stage 3
25 28 36
7 36 74
3 15 11
Total sessions
Total time (minutes)
1 5 7
30 54 96
Taylor, and Mike scored 3621 pixels, 3591 pixels, and 3969 pixels, respectively, during the post-assessment. The bottom three panels show the results of the number of trials in which the experimenters provided prompts. At the pre-assessment none of the participants could move the mouse appropriately and pressing the mouse button was not functional for pointing (e.g., they would repeatedly click the mouse without moving it or observing the on-screen cursor). For example, Beth did not move the mouse at all and Taylor and Mike sometimes lifted the mouse off the table surface. The experimenters provided prompts, especially physical prompts, at almost all trials during the pre-assessment. They did not have to provide any prompts at the post-assessment. 2.2. Beth Table 2 shows the total number of trials required to meet the criteria at each stage of training. Beth did not require any prompts to progress through the stages. The computer program was enough to shape her behaviors. The experimenters only changed the computer setting and provided praise statements. She started Stage 1 with a pixel detection length of 1 pixel. The detection length increased from 1 to 25 pixels. She finished Stage 1 in 25 trials. At Stage 2, she required five trials to meet the criteria for Stage 2-1 and two trials to meet the criteria for Stage 2-3. Stage 2-2 training was not necessary for Beth. At Stage 3, she required three trials to meet the criteria. She completed all stages in one 30-min session. 2.3. Taylor Taylor required five sessions to finish all stages of Programmed Instruction. The total time required to complete five sessions was 54 min. At the first session, he began at Stage 1 and made steady progress. The experimenters did not provide any prompts at Stage 1. The total number of trials required to finish Stage 1 was 28. He required two trials to meet the criteria when the pixel detection length was 21 pixels. He finished Stage 1 when the detection length was 27 pixels. After he finished Stage 1, he was engaged in Stage 2-1 for two trials. Then the first session was terminated. At the beginning of the second session, Taylor remained in Stage 1 as a ‘‘warm-up’’ for one trial with a pixel detection length of 25 pixels. After the warm-up, he moved to Stage 21 for five trials with the normal cursor size. The experimenters used physical prompts, verbal prompts, and gesturing to the screen during all of the trials, however, he did not meet the criteria for Stage 2-1. Therefore, the cursor size was enlarged (Stage 2-2) at the third
186
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
session. He engaged the task with the largest cursor (Size 10) in 14 trials after the same warm-up of the second session. He used a Macally imouse1 from the second trial to the last trial. The experimenters continued to use the same prompts during this session. At the beginning of the fourth session, he remained at Stage 2-2 for two trials with the largest cursor size (Size 10). Then the cursor size was gradually reduced from Size 9 to Size 1. He engaged in one trial at each cursor size, but required two trials at Size 1 (normal cursor size) to meet criteria. At Stage 2-3, he required three trials to meet criteria. The total number of trials required to finish Stage 2 was 36. The total number of trials required to finish Stage 3 was 15, the first 8 trials of which required the experimenter to provide prompts. He finished Stage 3 in the fifth session. 2.4. Mike Mike finished all stages of intervention in seven sessions totaling 96 min. In Stage 1, he spent one trial at the 1 pixel to 19 pixels detection length. He spent two trials at a detection length of 20 pixels and three trials at a detection length of 25 pixels. The experimenters provided physical prompts during some trials. Because the length of the mouse movement was very short, the detection length changed to 128 pixels. This change worked well for Mike. He met the criteria after eight trials at a detection length of 128 pixels. The total number of trials required to finish Stage 1 was 36. At the beginning of the second session, he was engaged in Stage 1 training for two trials as a warm-up. The detection length of the first and second trial was 25 pixels and 128 pixels, respectively. After the warm-up, he remained at Stage 2-1 for two trials with the normal cursor size. The experimenters used physical prompts, verbal prompts, and gesturing to the screen. He did not meet the criteria for Stage 2-1. The cursor size was enlarged (Stage 2-2) in the third session. He completed the task with the largest cursor (Size 10) for 18 trials after the same warm-up phase that occurred before the second session. A touch panel was used at trials 11, 12, and 13. During these trials, Mike pointed at the black rectangles with his index finger instead of using the mouse. He used a Macally imouse1 in the 17th and 18th trials. At the beginning of the fourth session, Mike spent five trials with the largest cursor size (Size 10). Then the cursor size changed. It was Size 9 at the sixth trial, Size 8 at trials 7 through 9, Size 7 at trials 10 and 11, and Size 6 at trials 12 and 13. Mike used an Apple USB Mouse in the first and second trial, a Logitech Cordless Wheel Mouse in the third trial, and a Macally imouse1 for the rest of the trials. The experimenter began keeping the mouse in the correct position, controlling the mouse cord, at the fifth session while other prompts remained in effect. Mike used an Apple USB Mouse during this session. He spent 10 trials with the largest cursor size, then the cursor size changed. He spent one trial at Size 9, Size 8, and Size 7, and two trials at Size 6 and Size 5. The session finished after he spent five trials at Size 4. The sixth session started with the largest cursor size (Size 10). Mike made steady progressed at this session. He spent two trials at Size 10 through Size 8, one trial at Size 7 through Size 3, two trials at Size 2, and three trials at Size 1 (normal), completing Stage 22. Mike met criteria in Stage 2-3 in three trials. The total number of trials required to finish Stage 2 was 74. He met the criteria of Stage 3 in 11 trials at the seventh session.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
187
3. Discussion This study was designed to teach three preschoolers with developmental disabilities to use a computer mouse to guide the on-screen cursor so they could point and click on items on a computer screen. The pre-assessment showed that none of the participants demonstrated this skill, however, they could press and release a mouse button. The postassessment showed that after intervention that included Programmed Instruction combined with experimenter-provided prompts, all participants learned the target skill. The intervention design was based on an analysis of the behavioral prerequisites for pointing. Stage 1 of the instruction was designed to teach participants to move the mouse. Stage 2 was designed to teach participants to move the cursor onto specific items on the screen. Stage 3 was designed to teach participants to click on specific items on the screen. The most difficult stage for the participants was Stage 2. Two of three participants (Taylor and Mike) required a number of trials, and physical, verbal, and gestural prompts to finish Stage 2. The behavior required in Stage 2 can be characterized as sensory–motor coordination. More specifically, it requires the coordination of hand and eye movements. It seems likely that their difficulty at Stage 2 was caused by the complexity of the hand–eye coordination required to complete the task. Individuals with developmental disabilities often show less sensory–motor coordination than typically developing individuals (e.g., Alcorn & Nicholson, 1972). We did not expect the level of difficulty we encountered before we started teaching hand–eye coordination skills associated with mouse control. However, as soon as we started the training, we realized that we had to analyze the hand–eye coordination prerequisites and create new procedures to shape and integrate each prerequisite. As the result of observation of the participants’ behavior at Stage 2, we analyzed the prerequisites into four components: (a) looking at the cursor, (b) looking at specific items on the screen where the cursor is to be moved, (c) moving the mouse to point the cursor at specific items, and (d) keeping the mouse in the correct position. The results suggest that the procedures used in this study, which were a combination of Programmed Instruction and experimenter-provided prompts, are effective to teach the hand–eye coordination required for pointing. Our next challenge will be to streamline the procedures. As mentioned above, we had to add new interventions depending on participants’ behavior during the programmed instruction, especially in Stage 2. We enlarged the cursor size in Stage 2 after two participants showed difficulty with the normal cursor size. Whenever the participants moved the mouse, the color of the cursor changed to promote observation of the cursor. We used a touch panel to give one participant further practice observing different elements of the computer screen. We changed the mouse when we found that it was difficult for the participants to keep the mouse positioned correctly. The intervention was sufficient to teach these participants basic mouse control skills and we believe the inductive approach we used in this experiment is highly compatible with the applied behavior analysis tradition. However, the experimental design and dependent variables used in the present study do not allow us to identify how each element of intervention influenced participants’ behavior. Some aspects of the intervention may have been unnecessary or inefficient. For example, did the enlarged cursor make the participants look at the cursor more often? We
188
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
might determine the effect of this element of the intervention procedure with the use of specialized equipment, such as a device that tracks eye movements. Future refinement of the intervention includes additional automation of instruction (e.g., Vargas & Vargas, 1992). Both the computer and the experimenters conducted the current procedures, and the procedures were complicated especially in Stage 2. However, there is every possibility to make the procedures automated. For example, the computer program appeared to shape the behavior of one participant without help from the experimenters; the experimenters provided words of encouragement and changed the computer settings but provided no prompts. With further procedural refinement, the program might be useful not only for children with developmental disabilities but also for adults with disabilities and typically developing children. Another topic for future consideration is to expand the current procedures into other mouse actions such as pressing and releasing the mouse button. The programmed instruction in the present study did not include procedures to teach pressing and releasing the mouse button because the participants demonstrated this skill during pre-assessment. However, there may be individuals who do not demonstrate skills in pressing and releasing the mouse button. Computer-based Programmed Instruction to teach this skill might be useful. Pressing and releasing the button consists of two components: (a) pressing and (b) releasing. A computer generally discriminates between pressing and releasing, and provides a user with different feedback. Pressing is often called MouseDown, and releasing is called MouseUp. In the present study we tried to isolate movement of the mouse from clicking the mouse. During instruction, the clicking function of the mouse was disabled so we could focus on the behavioral deficits: mouse movement and hand–eye coordination. When the mouse was enabled again at the end of the study the participants easily integrated the behavior they had (clicking) with the behavior they acquired through intervention (movement and hand–eye coordination). One idea to shape pressing and releasing is to isolate the clicking skill by fixing the mouse position. With the participants unable to move the mouse it might be easier to teach MouseUp and MouseDown functions. Then, as demonstrated in this study, the skills can be integrated. There are other essential mouse actions such as dragging and double-clicking. These skills are required for many advanced computer tasks and commonly used preschool software. Developing training programs to teach these skills would create more opportunities for people who have difficulty using a computer mouse, and it would allow them to access the growing number of well-designed educational program available to computer users.
Acknowledgements The first author served as a Research Fellow of the Japan Society for the Promotion of Science when this research was conducted. This research was supported in part by the Ministry of Education of Japan Society (Grant-in-Aid for JSPS Fellows H11-00148). We would like to thank Tina Covington, Program Supervisor at Hawthorne Country Day School, for her support in finishing this manuscript. Thanks also to Ann Marie Babcock, Education Coordinator at Hawthorne Country Day School, for her help in preparing this manuscript.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
189
References Alcorn, C. L., & Nicholson, C. L. (1972). Validity of the Slosson Drawing Coordination Test with adolescents of below-average ability. Perceptual and Motor Skills, 34, 261–262. Dube, W. V., Iennaco, F. M., Rocco, F. J., Kledaras, J. B., & McIlvane, W. J. (1992). Microcomputer-based programmed instruction in identity matching to sample for persons with severe disabilities. Journal of Behavioral Education, 2, 29–51. Dube, W. V., McDonald, S. J., & McIlvane, W. J. (1991). Constructed-response matching to sample and spelling instruction. Journal of Applied Behavior Analysis, 24, 305–317. Durfee, J. L., & Billingsley, F. F. (1998). A comparison of two computer input devices for uppercase letter matching. The American Journal of Occupational Therapy, 53, 214–220. Joiner, R., Messer, D., Light, P., & Littleton, K. (1998). It is best to point for young children: A comparison of children’s pointing and dragging. Computers in Human Behavior, 14, 513–529. King, J. (1992). Preschooler’s use of microcomputers and input devices. Journal of Educational Computing Research, 8, 451–468. Lancioni, G. E., & Boelens, H. (1996). Teaching students with mental retardation and other disabilities to make simple drawings through a computer system and special catds. Perceptual and Motor Skills, 83, 401–402. Lane, S. D., & Critchfield, T. S. (1998). Classification of vowels and consonants by individuals with moderate mental retardation: Development of arbitrary relation via match-to-sample training with compound stimuli. Journal of Applied Behavior Analysis, 31, 21–41. Leung, J. (1994). Teaching simple addition to children with mental retardation using a microcomputer. Journal of Behavioral Education, 4, 355–367. MacKenzie, I. S. (1992). Fitts’ Law as a research and design tool in human-computer interaction. HumanComputer Interaction, 7, 91–139. Mastropieri, M. A., Scruggs, T. E., & Shiah, R. (1997). Can computers teach problem-solving strategies to students with mild mental retardation? Remedial and Special Education, 18, 157–165. Merzenich, M. M., Jenkins, W. M., Johnston, P., Schreiner, C., Miller, S. L., & Tallal, P. (1996). Temporal processing deficits of language-learning impaired children ameliorated by training. Science, 271, 77–81. Missiuna, C. (1994). Motor skill acquisition in children with developmental coordination disorder. Adapted Physical Activity Quarterly, 11, 214–235. Neef, N. A., Bicard, D. F., & Endo, S. (2001). Assessment of impulsivity and the development of self-control in students with attention deficit hyperactivity disorder. Journal of Applied Behavior Analysis, 34, 397–408. Saunders, K. J., Johnston, M. D., Tompkins, B. F., Dutcher, D. L., & Williams, D. C. (1997). Generalized identity matching of two-dimensional forms by individuals with moderate to profound mental retardation. American Journal on Mental Retardation, 102, 285–291. Shimizu, H., & Yamamoto, J. (2000, May). Developing a computer program for teaching students with developmental disabilities to use a mouse. Poster session presented at the 26th annual convention of the Association for Behavior Analysis, Washington, DC. Stromer, R., Mackay, H. A., Howell, S. R., & McVay, A. A. (1996). Teaching computer-based spelling to individuals with developmental and hearing disabilities: Transfer of stimulus control to writing tasks. Journal of Applied Behavior Analysis, 29, 25–42. Stronmen, E. F., Revelle, G. L., Medoff, L. M., & Razavi, S. (1996). Slow and steady wins the race? Three-yearold children and pointing device use. Behavior and Information Technology, 15, 57–64. Tallal, P., Miller, S. L., Bedi, G., Byma, G., Wang, X., Nagarajan, S. S., et al. (1996). Language comprehension in language-learning impaired children improved with acoustically modified speech. Science, 271, 81–84. Touchette, P. E., & Howard, J. S. (1984). Errorless learning: Reinforcement contingencies and stimulus control transfer in delayed prompting. Journal of Applied Behavior Analysis, 17, 175–188. Vargas, E. A., & Vargas, J. S. (1992). Programmed instruction and teaching machines. In R. P. West & L. A. Hamerlynck (Eds.), Designs for excellence in education: The legacy of B.F. Skinner (pp. 33–69). CO: Sopris West.
Programmed Instruction to teach pointing with a computer mouse in preschoolers with developmental disabilities Hirofumi Shimizu a,b,*, Christopher S. McDonough b a
b
Meisei University, Tokyo, Japan Hawthorne Country Day School, 5 Bradhurst Avenue, Hawthorne, NY 10532, USA
Received 30 September 2004; received in revised form 20 January 2005; accepted 23 January 2005
Abstract Programmed Instruction combined with experimenter-provided prompts (physical, verbal, and gesturing) was used to teach pointing with a computer mouse. Three preschoolers who scored at least 1 year below their chronological age levels participated. During the pre-assessment, none of the participants demonstrated pointing. However, they could press and release the mouse button. Programmed Instruction consisted of three stages, based on an analysis of the behavioral prerequisites for pointing. Stage 1 was designed to teach participants to move the mouse. Stage 2 was designed to teach participants to move the on-screen cursor onto specific items on the screen. Stage 3 was designed to teach participants to click on specific items on the screen. Experimenter-provided prompts were used to facilitate skill acquisition at each stage. The post-assessment showed that all participants learned pointing after intervention. The intervention package consisting of Programmed Instruction and experimenter-provided prompts was effective for teaching the hand–eye coordination required for pointing. # 2005 Elsevier Ltd. All rights reserved. Keywords: Computer-based instruction; Developmental disabilities; Mouse; Programmed Instruction; Preschooler
* Corresponding author. Tel.: +1 914 592 8526x166; fax: +1 914 592 3227. E-mail address: [email protected] (H. Shimizu). 0891-4222/$ – see front matter # 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.ridd.2005.01.001
176
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
Computer-based instruction is widely used in special education (e.g., Lancioni & Boelens, 1996; Leung, 1994; Mastropieri, Scruggs, & Shiah, 1997; Merzenich et al., 1996; Tallal et al., 1996), including the field of Applied Behavior Analysis (e.g., Dube, Iennaco, Rocco, Kledaras, & McIlvane, 1992; Dube, McDonald, & McIlvane, 1991; Lane & Critchfield, 1998; Neef, Bicard, & Endo, 2001; Saunders, Johnston, Tompkins, Dutcher, & Williams, 1997; Stromer, Mackay, Howell, & McVay, 1996). In the previous reports, many researchers have used a touch panel as a computer-input device. Although there might be other reasons for researchers to use a touch panel (e.g., automated data collection for selection-based response classes), special education teachers may sometimes encounter difficulty teaching their students to use a mouse. For such students, unless a computer is equipped with a specialized alternative input device such as a touch panel, their access to the growing number of well-designed educational programs available to computer users is limited. A mouse is one of the most familiar input devices. This device can be used for almost all computers. The movement of an arrow-shaped cursor on the computer screen is achieved by moving the mouse. People can interact with a computer by pressing and releasing the button of the mouse. The two basic mouse actions most commonly used are pointing and dragging. For pointing, people move the mouse in order to move a cursor on the computer screen. When the cursor is pointed at a specific item such as an icon, picture, or text, they press and release the mouse button to send functional commands to the computer. Pressing and releasing the mouse button to send functional commands is often called click or clicking. In dragging, the mouse button is pressed and held down when the cursor is on a specific item on the screen and then the mouse is moved while the button remains held down. When the cursor reaches another specific item the mouse button is released sending a command to the computer. Dragging is often used to draw, move, or highlight objects. Previous studies have shown that normal four-year-old preschoolers demonstrate pointing and dragging (e.g., King, 1992). Even three-year-old children show the ability to use a mouse (e.g., Stronmen, Revelle, Medoff, & Razavi, 1996). Pointing is easier than dragging for both children (e.g., Joiner, Messer, Light, & Littleton, 1998) and adults (e.g., Mackenzie, 1992). Some studies have indicated that children with developmental disabilities have the ability to learn pointing (e.g., Durfee & Billingsley, 1998; Missiuna, 1994; Shimizu & Yamamoto, 2000). For example, Durfee and Billingsley (1998) reported that a mouse was a more effective interface device than a touch panel for a boy with spastic quadriplegic cerebral palsy. Shimizu and Yamamoto (2000) assessed the pointing skill of 16 students with severe and moderate mental retardation ranging in age from 7 to 12 years old. Eight students demonstrated pointing without any problems, four students demonstrated gradual improvement throughout the assessment and four students could not use a mouse. In addition to these findings, we have found in our laboratory that many preschoolers who are identified as ‘‘preschoolers with disabilities’’ demonstrate the ability to use a mouse. However, some do not. The number of children who are able to use a mouse might increase if the teaching process is highly controlled with the application of behavioral technology. The purpose of the present study was to develop and evaluate an instructional sequence (intervention) to teach pointing with a computer mouse to three preschoolers with developmental disabilities. The strategy had three characteristics: (a) breaking the terminal
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
177
skill into several component skills, (b) shaping each component and systematically integrating them, and (c) errorless teaching throughout instruction. The intervention consisted of Programmed Instruction and experimenter-provided prompts (physical, verbal, and gesturing). The terminal skill, pointing, was broken down into four component skills: (a) pressing and releasing the mouse button, (b) moving the mouse, (c) pointing the cursor at specific items on the screen, and (d) clicking on specific items on the screen. Since all of the participants demonstrated the first component skill (pressing and releasing the mouse button) during pre-assessment, instruction was not provided for this component skill. The instruction consisted of three stages to shape each of the other component skills and to systematically integrate them. The first stage of training was designed to shape movement of the mouse. The second stage was designed to teach pointing the cursor at specific items on the screen, a skill requiring precise hand–eye coordination. Because of the complexity of this skill it was broken down into four sub-components: (a) looking at the cursor, (b) looking at specific items on the screen where the cursor was to be moved, (c) moving the mouse to point the cursor at specific items, and (d) keeping the mouse in the correct position (i.e., facing the computer). The third stage was designed to teach participants to click on the specific items on the screen. To promote errorless learning, Programmed Instruction was supported by experimenter-provided prompts.
1. Method 1.1. Participants Three 4-year-old children with developmental disabilities participated: Beth, Mike and Taylor. Each participant was identified as a ‘‘preschooler with a disability’’ by his or her school district’s Committee on Preschool Special Education. The children were enrolled in the special education preschool program where this study was conducted. All participants scored at least 1 year below their chronological age level on standardized assessments. Their dominant arms were right. These children could press and release the mouse button before the training, but could not use the mouse for pointing. 1.2. Apparatus and setting Instruction occurred in brief sessions (average 14 min) in a quiet room at the preschool one to four times per week. The computer, an Apple iMac1, presented stimuli and recorded data while the experimenters provided and recorded prompts. The computer software for assessment was developed with Apple Hypercard1 2.2, while the software for training was created with Macromedia Director1 8.0. The computer input device was an Apple USB Mouse. The terminal goal was for participants to acquire pointing using this mouse. Participant Mike used two other mice in the process (Stage 2-2). One mouse was a Macally imouse1, and the other was a Logitech Cordless Wheel. Taylor also used a Macally imouse1 during Stage 2-2. In addition to the mice, Mike used a touch panel (Troll Touch TouchStar) at the beginning of Stage 2-2.
178
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
The computer was placed on a 1 m 1 m table. The participants sat on a chair, facing the computer screen. There was space (30 cm 60 cm) on the table between the participant and the computer, and participants manipulated the mouse in this area. The keyboard was located on the left side of the computer out of the participant’s reach. The two experimenters sat behind and to either side of the participant. 1.3. Assessment for pointing 1.3.1. Procedure Assessment occurred before and after training. The assessment tool used was the computer software reported by Shimizu and Yamamoto (2000). Fig. 1 illustrates the computer screens of the assessment. Assessment began with a click on the ‘‘START’’ button by the experimenter (Fig. 1A). The computer presented the auditory instruction: ‘‘What’s hiding behind the squares?’’ The screen was filled with various sizes of 15 black rectangles (Fig. 1B) and a picture was ‘‘hiding’’ behind the rectangles. Although there were small gaps between the rectangles, the picture was not in view. It resembled a puzzle placed facedown. As the participants pointed the on-screen cursor at the rectangles and pressed the mouse button (clicked on the rectangles), the clicked rectangle disappeared with a brief chime, revealing part of the picture (Fig. 1C and D). After all of the rectangles had been clicked the whole picture appeared with a brief musical fanfare (Fig. 1E). The opportunity to click on one rectangle was defined as one trial. Clicking on the 15 rectangles was defined as one block. After one block was completed, the next block began with the presentation of 15 more black rectangles and the auditory instruction: ‘‘What will you find next?’’ Five consecutive blocks were conducted during the pre-assessment and
Fig. 1. The sample computer screens for the pre-assessment and the post-assessment phases. Panel A: The assessment began with the ‘‘START’’ button pressed. Panel B: At the beginning of one block, 15 black rectangles appeared on the screen. A picture was hidden behind the rectangles. Panels C and D: As participants clicked on the rectangles, the clicked rectangle disappeared, revealing part of the picture. Panel E: As they clicked on all the rectangles, the whole picture appeared. Panel F: The assessment ended.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
179
post-assessment. The picture behind the rectangles changed from block to block. After five blocks were completed, the word ‘‘END’’ appeared on the screen with a short jingle (Fig. 1F). The experimenters used a delayed prompting procedure (e.g., Touchette & Howard, 1984) during assessment. If the participants did not click on a rectangle after 5 s, one of the experimenters guided their performance with one of three kinds of prompts. (a) Physical prompt: An experimenter helped the participant move the mouse or press the mouse button (or both), holding the participant’s right hand or arm. (b) Verbal prompt: An experimenter said, ‘‘Move it’’ or ‘‘Push the button’’. (c) Gesturing to the screen: An experimenter pointed his index finger at the rectangles on the computer screen. 1.3.2. Measurement Four measures were used to assess pointing. (a) Total number of clicks per block: One click was defined as one press of the mouse button. (b) Total time per block: This measure was defined as the length of time it took for the participant to click on all 15 black rectangles. Time recording began when the rectangles were present on the screen and stopped after the last rectangle was clicked. (c) Total length of on-screen cursor movement per block: This measure was the combined length of all cursor movement during the block. The measurement unit was recorded in pixels. A pixel is a unit used to describe the resolution of a computer screen. One pixel is the smallest element. For example, 640 480 resolution means that the screen consists of the 640 pixels in width and 480 pixels in height. The resolution of the computer display in this experiment was 640 480. If a participant moved the cursor from one side of the screen to the other, the computer recorded the length as 640 pixels. All cursor movement during the block was recorded and summed to determine the total length of on-screen cursor movement. (d) Total number of trials per block in which the experimenters provided prompts was the fourth measure. 1.3.3. Interobserver agreement Interobserver agreement related to prompting was calculated between the two experimenters. They independently recorded the incidence of prompting during assessment sessions in real time. Agreement was calculated by dividing agreements by agreements plus disagreements and multiplying by 100%. Interobserver agreement was 96%. The computer automatically recorded other dependent measures. 1.4. Programmed instruction procedures Table 1 illustrates a training goal and an intervention package at each stage of the programmed instruction. The programmed instruction consisted of three stages and was designed to teach and integrate: (a) moving the mouse; (b) pointing the cursor at specific items on the screen; and (c) clicking on specific items on the screen. 1.5. Stage 1: moving the mouse Stage 1 was designed to teach moving a mouse. All three participants pressed and released the mouse button frequently during pre-assessment and the initial teaching of
180
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
Table 1 Training goal and experimenter-provided prompts Stage
Goal
Experimenter-provided prompts
1
Moving the mouse
1. Physical prompts to help move the mouse
2
Pointing the cursor at specific items on the screen
1. Physical prompts to help move the mouse 2. Verbal prompts (e.g., ‘‘move the mouse’’ and ‘‘look at this’’) 3. Gesturing to the screen: Pointing trainer’s index finger to the cursor or rectangles 4. Keeping the mouse in the correct position (this was used only in Stage 2-2)
3
Clicking on specific items on the screen
1. Physical prompt to help move the mouse and press the mouse button 2. Verbal prompts (e.g., ‘‘Press the button’’, ‘‘Push the button’’)
Stage 1. Since this appeared to interfere with shaping movement of the mouse, the mouse button was disabled so the participants could not press and release the button. A completely black screen was presented at the beginning of each trial (Fig. 2A). The screen consisted of 25 black rectangles, all the same size (128 96 pixels). A picture was hidden behind the rectangles. If the participants moved the mouse, three forms of feedback followed the movement (Fig. 2B): (a) one rectangle disappeared and the part of the picture corresponding to that rectangle was revealed, (b) the color of the remaining rectangles
Fig. 2. The sample computer screens for Programmed Instruction. Panel A (Stages 1 and 2): At the beginning of the trial, a completely black screen was presented. The screen consisted of 25 black rectangles, all the same size. A picture was hidden behind the rectangles. Panel B (Stage 1): As participants moved the mouse in Stage 1, the rectangles disappeared and the color of the remaining rectangles changed with a brief sound. Panels C and D (Stage 1): The more participants moved the mouse, the more rectangles disappeared, the more color changes occurred in the remaining rectangles, and the more the participants heard the brief sound. Panel E (Stage 2-2): The size of the cursor was enlarged. The participants were required to point the cursor at rectangles. Panel F (Stage 22): Whenever the cursor moved for 1 pixel on the screen, the color of the cursor randomly changed. When the cursor was pointed at rectangles they disappeared with a brief chime.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
181
changed, and (c) a brief sound was presented. The computer randomly determined which rectangle disappeared and the color of the remaining rectangles. The more often the participants moved the mouse, the more rectangles disappeared, the more color changes occurred in the remaining rectangles, and the more participants heard the brief sound (Fig. 2C and D). When all the rectangles disappeared, the entire picture was revealed with a brief musical fanfare. Clearing all 25 rectangles to reveal the picture was defined as one trial. The detection of the mouse movement by the computer depended on the length of the on-screen cursor movement measured in pixels. At the beginning of Stage 1, the detected length of pixels was set to 1 pixel. One pixel movement of the on-screen cursor represented approximately 0.6 mm movement of the mouse on the table. In other words, when the participant moved the mouse on the table for 0.6 mm, the cursor moved for 1 pixel on the screen, and one black rectangle disappeared. Therefore, at the beginning of Stage 1, one trial was defined as the clearing of all 25 rectangles, which was equivalent to moving the cursor on the screen for 25 pixels and the mouse for approximately 15 mm. The detected length increased pixel by pixel, depending on the participant’s progress. For example, if the participant met criteria at the trial with a detection length of 1 pixel, the experimenters set the detection length to 2 pixels for the next trial. The criteria to increase the pixel detection length were that (a) the participant moved the mouse without any prompts, and (b) the total time to finish the trial was less than 25 s. As illustrated in Table 1, the experimenter provided physical prompts at Stage 1. If the participant could not move the mouse, the experimenters prompted the participant to move the mouse, holding their right hand or arm. Whenever one trial finished without any prompts, the experimenters applauded the participant with words or phrases (e.g., ‘‘wonderful’’, ‘‘excellent’’, and ‘‘good job’’). Stage 1 ended when the participant met the criteria and they spontaneously moved the mouse on the table more than 30 cm. 1.6. Stage 2: pointing the cursor at specific items on the screen Stage 2 was designed to teach pointing the on-screen cursor at specific items on the screen. This skill requires relatively complex hand–eye coordination. It requires (a) looking at the cursor, (b) looking at specific items on the screen where the cursor is to be moved, (c) moving the mouse to point the cursor at specific items, and (d) keeping the mouse in the correct position, which is pointed toward the computer. Stage 2 was divided into three parts: 2-1, 2-2, and 2-3. Stage 2-2 was developed in response to two participants failing to meet criteria at Stage 2-1 (described below). The mouse button remained disabled throughout Stage 2 so that the participant could not press and release the button. 1.6.1. Stage 2-1 Stage 2-1 began like Stage 1 with the same black screen consisting of 25 black rectangles. The main difference between Stages 1 and 2-1 was that in Stage 2-1 the participants were required to point the cursor at each of the rectangles on the screen. In Stage 1, it did not matter if the cursor pointed to each rectangle or not, only that it moved a predefined number of pixels. In Stage 2-1, when the cursor was pointed at a rectangle, that rectangle disappeared with a brief chime and a portion of the picture appeared in its place.
182
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
The other difference between Stages 1 and 2-1 was that the color of the remaining rectangles did not change with cursor movement. The rectangles always remained black. In Stage 2-1 one trial was defined as moving the cursor so it pointed at each of the 25 rectangles on the screen. When necessary, the experimenters used prompts during Stage 2-1 as described in Table 1. If the participants did not move the mouse after 5 s, one of the experimenters prompted the participants to move the mouse either physically or verbally. Furthermore, if the participants did not look at the cursor or rectangles, one of the experimenters pointed their index finger to the cursor or rectangles and said, ‘‘Look at this’’. When all 25 rectangles disappeared or one trial was completed without any prompts, the experimenters applauded participants’ performance with words or phrases. The criteria for Stage 2-1 were that (a) the participants pointed the cursor at all rectangles without any prompts in three consecutive trials, and (b) the total time to finish those trials was less than 15 s. Only Beth could meet these criteria. 1.6.2. Stage 2-2 Stage 2-2 was developed because Taylor and Mike failed to meet Stage 2-1 criteria. During Stage 2-1 Taylor and Mike (especially Mike) were observed to have trouble controlling the mouse. That is, the mouse would become rotated to a sideways or upside down position. When the mouse is out of position relative to the computer screen it becomes difficult to operate because the cursor movement is not reflective of the mouse movement. In addition, Taylor and Mike required frequent gestural and verbal prompts to continue looking at the on-screen cursor. This could indicate that they did not consistently observe the on-screen cursor during Stage 2-1, an observation without which they had no chance at acquiring the target skill. Therefore, to assist Taylor and Mike with observation of the on-screen cursor and control of the mouse, Stage 2-2 was developed. In Stage 2-2 the initial size of the on-screen cursor was about 800% (Fig. 2E and F) of normal size (11 16 pixels, which is the size used throughout the experiment). In addition, the computer was programmed to randomly change the color of this enlarged cursor whenever it moved 1 pixel on the screen. The cursor size gradually decreased across 10 steps depending on the participant’s progress. The criteria to change the cursor size were that (1) the participant pointed the cursor at all rectangles without prompts, and (2) the total time to finish the trial was less than 20 s. Other Programmed Instruction features of Stage 2-2 were identical to Stage 2-1. Two things were done in an effort to help Taylor and Mike control the mouse. First, the mouse cord was shortened so that moving the mouse out of position became difficult. This was done only for Mike. The length of the mouse cord also was manually controlled by keeping it without slack while Mike was moving the mouse. Second, two different mice were used with Mike (a Macally imouse1, which has an oval shape unlike the round Apple USB Mouse, and a Logitech Cordless Wheel Mouse) and one with Taylor (the Macally imouse1). In addition, the prompts used in Stage 2-1 remained in effect. Mike was allowed to use a touch screen for three trials in Stage 2-2 in an effort to help him look at the on-screen cursor and to experience independent success with pointing to
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
183
rectangles on the computer screen. The hand–eye coordination required for pointing to items on the computer screen with one’s finger is not as difficult as using the mouse to control the on-screen cursor. When using the touch screen, the enlarged on-screen cursor moved to the point where Mike’s finger contacted the screen. 1.6.3. Stage 2-3 Stage 2-3 was the same as Stage 2-1 except that the screen used during 2-3 was the same as the screen used during assessment (Fig. 1B). If the participants pointed the cursor at the rectangles they disappeared, with a brief chime, and part of the picture was revealed. The criteria to complete Stage 2-3 were that (1) the participant pointed the cursor at all rectangles without any prompts in two consecutive trials, and (2) the total time to finish those trials was less than 15 s. 1.7. Stage 3: teaching clicking on specific items on the screen Stage 3 was designed to teach clicking on specific items on the computer screen. The mouse button was enabled at this stage. The computer program at Stage 3 was the same one used for assessment, however, the program only recorded duration. The participants were required to point the cursor at rectangles, and to click on them. The experimenters prompted them to point the cursor at rectangles, and click on rectangles, using physical prompts or verbal prompts (e.g., ‘‘Push the button’’, ‘‘Press the button’’). When rectangles disappeared or one trial was completed without any prompts, the experimenters praised the participant. The criteria to complete Stage 3 were that (1) the participants clicked on all 15 rectangles without any prompts in three consecutive trials, and (2) the total time to finish those trials was less than 30 s.
2. Results 2.1. Assessment Fig. 3 shows the results of the pre-assessment and the post-assessment for each participant across four dependent measures. Pre-assessment scores and post-assessment scores are shown as gray and white bars, respectively. Five blocks were conducted during pre-assessment and post-assessment and each bar represents the average of five blocks. The Y error bars (vertical hairlines stuck into the top of each bar) represent the range of the data in the five blocks. The top three panels of Fig. 3 show the results of the total number of clicks per block. Fifteen was the minimum number of clicks required to clear the rectangles and reveal the entire picture because there were 15 rectangles. All participants could press and release the mouse button at the pre-assessment. Taylor and Mike showed very high numbers of clicks at the pre-assessment (ranged from 47 to 61 for Taylor and from 94 to 138 for Mike) while the number of clicks for Beth ranged from 22 to 35. Performance greatly improved after intervention. The number of clicks at the post-assessment ranged from 15 to 16 for Beth, 18 to 21 for Taylor, and 19 to 22 for Mike.
184
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
Fig. 3. The results of the pre-assessment and the post-assessment phases. Pre-assessment scores and postassessment scores are shown as gray and white bars, respectively. Five blocks were conducted at both preassessment and post-assessment and each bar represents the average of five blocks. The Y error bars (vertical hairlines stuck into the top of each bars) represent the range of the data in the five blocks.
The second three panels are the result of the total time per block. All the participants took about 2 min to finish one block at the pre-assessment. The speed was drastically improved after intervention. They each took about 20 s to finish one block. The third three panels show the results of the total length of cursor movement per block. The data from a previous study indicate that the length of cursor movement is around 3800 pixels on average, if students with disabilities demonstrate appropriate pointing (Shimizu & Yamamoto, 2000). Beth and Mike scored 1416 pixels and 2807 pixels, respectively, on average at the pre-assessment, while Taylor scored 3591 pixels. Beth,
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
185
Table 2 The total number of trials to meet criteria at each stage, total number of sessions, and total time for intervention Participants
Beth Taylor Mike
The total number of trials Stage 1
Stage 2
Stage 3
25 28 36
7 36 74
3 15 11
Total sessions
Total time (minutes)
1 5 7
30 54 96
Taylor, and Mike scored 3621 pixels, 3591 pixels, and 3969 pixels, respectively, during the post-assessment. The bottom three panels show the results of the number of trials in which the experimenters provided prompts. At the pre-assessment none of the participants could move the mouse appropriately and pressing the mouse button was not functional for pointing (e.g., they would repeatedly click the mouse without moving it or observing the on-screen cursor). For example, Beth did not move the mouse at all and Taylor and Mike sometimes lifted the mouse off the table surface. The experimenters provided prompts, especially physical prompts, at almost all trials during the pre-assessment. They did not have to provide any prompts at the post-assessment. 2.2. Beth Table 2 shows the total number of trials required to meet the criteria at each stage of training. Beth did not require any prompts to progress through the stages. The computer program was enough to shape her behaviors. The experimenters only changed the computer setting and provided praise statements. She started Stage 1 with a pixel detection length of 1 pixel. The detection length increased from 1 to 25 pixels. She finished Stage 1 in 25 trials. At Stage 2, she required five trials to meet the criteria for Stage 2-1 and two trials to meet the criteria for Stage 2-3. Stage 2-2 training was not necessary for Beth. At Stage 3, she required three trials to meet the criteria. She completed all stages in one 30-min session. 2.3. Taylor Taylor required five sessions to finish all stages of Programmed Instruction. The total time required to complete five sessions was 54 min. At the first session, he began at Stage 1 and made steady progress. The experimenters did not provide any prompts at Stage 1. The total number of trials required to finish Stage 1 was 28. He required two trials to meet the criteria when the pixel detection length was 21 pixels. He finished Stage 1 when the detection length was 27 pixels. After he finished Stage 1, he was engaged in Stage 2-1 for two trials. Then the first session was terminated. At the beginning of the second session, Taylor remained in Stage 1 as a ‘‘warm-up’’ for one trial with a pixel detection length of 25 pixels. After the warm-up, he moved to Stage 21 for five trials with the normal cursor size. The experimenters used physical prompts, verbal prompts, and gesturing to the screen during all of the trials, however, he did not meet the criteria for Stage 2-1. Therefore, the cursor size was enlarged (Stage 2-2) at the third
186
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
session. He engaged the task with the largest cursor (Size 10) in 14 trials after the same warm-up of the second session. He used a Macally imouse1 from the second trial to the last trial. The experimenters continued to use the same prompts during this session. At the beginning of the fourth session, he remained at Stage 2-2 for two trials with the largest cursor size (Size 10). Then the cursor size was gradually reduced from Size 9 to Size 1. He engaged in one trial at each cursor size, but required two trials at Size 1 (normal cursor size) to meet criteria. At Stage 2-3, he required three trials to meet criteria. The total number of trials required to finish Stage 2 was 36. The total number of trials required to finish Stage 3 was 15, the first 8 trials of which required the experimenter to provide prompts. He finished Stage 3 in the fifth session. 2.4. Mike Mike finished all stages of intervention in seven sessions totaling 96 min. In Stage 1, he spent one trial at the 1 pixel to 19 pixels detection length. He spent two trials at a detection length of 20 pixels and three trials at a detection length of 25 pixels. The experimenters provided physical prompts during some trials. Because the length of the mouse movement was very short, the detection length changed to 128 pixels. This change worked well for Mike. He met the criteria after eight trials at a detection length of 128 pixels. The total number of trials required to finish Stage 1 was 36. At the beginning of the second session, he was engaged in Stage 1 training for two trials as a warm-up. The detection length of the first and second trial was 25 pixels and 128 pixels, respectively. After the warm-up, he remained at Stage 2-1 for two trials with the normal cursor size. The experimenters used physical prompts, verbal prompts, and gesturing to the screen. He did not meet the criteria for Stage 2-1. The cursor size was enlarged (Stage 2-2) in the third session. He completed the task with the largest cursor (Size 10) for 18 trials after the same warm-up phase that occurred before the second session. A touch panel was used at trials 11, 12, and 13. During these trials, Mike pointed at the black rectangles with his index finger instead of using the mouse. He used a Macally imouse1 in the 17th and 18th trials. At the beginning of the fourth session, Mike spent five trials with the largest cursor size (Size 10). Then the cursor size changed. It was Size 9 at the sixth trial, Size 8 at trials 7 through 9, Size 7 at trials 10 and 11, and Size 6 at trials 12 and 13. Mike used an Apple USB Mouse in the first and second trial, a Logitech Cordless Wheel Mouse in the third trial, and a Macally imouse1 for the rest of the trials. The experimenter began keeping the mouse in the correct position, controlling the mouse cord, at the fifth session while other prompts remained in effect. Mike used an Apple USB Mouse during this session. He spent 10 trials with the largest cursor size, then the cursor size changed. He spent one trial at Size 9, Size 8, and Size 7, and two trials at Size 6 and Size 5. The session finished after he spent five trials at Size 4. The sixth session started with the largest cursor size (Size 10). Mike made steady progressed at this session. He spent two trials at Size 10 through Size 8, one trial at Size 7 through Size 3, two trials at Size 2, and three trials at Size 1 (normal), completing Stage 22. Mike met criteria in Stage 2-3 in three trials. The total number of trials required to finish Stage 2 was 74. He met the criteria of Stage 3 in 11 trials at the seventh session.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
187
3. Discussion This study was designed to teach three preschoolers with developmental disabilities to use a computer mouse to guide the on-screen cursor so they could point and click on items on a computer screen. The pre-assessment showed that none of the participants demonstrated this skill, however, they could press and release a mouse button. The postassessment showed that after intervention that included Programmed Instruction combined with experimenter-provided prompts, all participants learned the target skill. The intervention design was based on an analysis of the behavioral prerequisites for pointing. Stage 1 of the instruction was designed to teach participants to move the mouse. Stage 2 was designed to teach participants to move the cursor onto specific items on the screen. Stage 3 was designed to teach participants to click on specific items on the screen. The most difficult stage for the participants was Stage 2. Two of three participants (Taylor and Mike) required a number of trials, and physical, verbal, and gestural prompts to finish Stage 2. The behavior required in Stage 2 can be characterized as sensory–motor coordination. More specifically, it requires the coordination of hand and eye movements. It seems likely that their difficulty at Stage 2 was caused by the complexity of the hand–eye coordination required to complete the task. Individuals with developmental disabilities often show less sensory–motor coordination than typically developing individuals (e.g., Alcorn & Nicholson, 1972). We did not expect the level of difficulty we encountered before we started teaching hand–eye coordination skills associated with mouse control. However, as soon as we started the training, we realized that we had to analyze the hand–eye coordination prerequisites and create new procedures to shape and integrate each prerequisite. As the result of observation of the participants’ behavior at Stage 2, we analyzed the prerequisites into four components: (a) looking at the cursor, (b) looking at specific items on the screen where the cursor is to be moved, (c) moving the mouse to point the cursor at specific items, and (d) keeping the mouse in the correct position. The results suggest that the procedures used in this study, which were a combination of Programmed Instruction and experimenter-provided prompts, are effective to teach the hand–eye coordination required for pointing. Our next challenge will be to streamline the procedures. As mentioned above, we had to add new interventions depending on participants’ behavior during the programmed instruction, especially in Stage 2. We enlarged the cursor size in Stage 2 after two participants showed difficulty with the normal cursor size. Whenever the participants moved the mouse, the color of the cursor changed to promote observation of the cursor. We used a touch panel to give one participant further practice observing different elements of the computer screen. We changed the mouse when we found that it was difficult for the participants to keep the mouse positioned correctly. The intervention was sufficient to teach these participants basic mouse control skills and we believe the inductive approach we used in this experiment is highly compatible with the applied behavior analysis tradition. However, the experimental design and dependent variables used in the present study do not allow us to identify how each element of intervention influenced participants’ behavior. Some aspects of the intervention may have been unnecessary or inefficient. For example, did the enlarged cursor make the participants look at the cursor more often? We
188
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
might determine the effect of this element of the intervention procedure with the use of specialized equipment, such as a device that tracks eye movements. Future refinement of the intervention includes additional automation of instruction (e.g., Vargas & Vargas, 1992). Both the computer and the experimenters conducted the current procedures, and the procedures were complicated especially in Stage 2. However, there is every possibility to make the procedures automated. For example, the computer program appeared to shape the behavior of one participant without help from the experimenters; the experimenters provided words of encouragement and changed the computer settings but provided no prompts. With further procedural refinement, the program might be useful not only for children with developmental disabilities but also for adults with disabilities and typically developing children. Another topic for future consideration is to expand the current procedures into other mouse actions such as pressing and releasing the mouse button. The programmed instruction in the present study did not include procedures to teach pressing and releasing the mouse button because the participants demonstrated this skill during pre-assessment. However, there may be individuals who do not demonstrate skills in pressing and releasing the mouse button. Computer-based Programmed Instruction to teach this skill might be useful. Pressing and releasing the button consists of two components: (a) pressing and (b) releasing. A computer generally discriminates between pressing and releasing, and provides a user with different feedback. Pressing is often called MouseDown, and releasing is called MouseUp. In the present study we tried to isolate movement of the mouse from clicking the mouse. During instruction, the clicking function of the mouse was disabled so we could focus on the behavioral deficits: mouse movement and hand–eye coordination. When the mouse was enabled again at the end of the study the participants easily integrated the behavior they had (clicking) with the behavior they acquired through intervention (movement and hand–eye coordination). One idea to shape pressing and releasing is to isolate the clicking skill by fixing the mouse position. With the participants unable to move the mouse it might be easier to teach MouseUp and MouseDown functions. Then, as demonstrated in this study, the skills can be integrated. There are other essential mouse actions such as dragging and double-clicking. These skills are required for many advanced computer tasks and commonly used preschool software. Developing training programs to teach these skills would create more opportunities for people who have difficulty using a computer mouse, and it would allow them to access the growing number of well-designed educational program available to computer users.
Acknowledgements The first author served as a Research Fellow of the Japan Society for the Promotion of Science when this research was conducted. This research was supported in part by the Ministry of Education of Japan Society (Grant-in-Aid for JSPS Fellows H11-00148). We would like to thank Tina Covington, Program Supervisor at Hawthorne Country Day School, for her support in finishing this manuscript. Thanks also to Ann Marie Babcock, Education Coordinator at Hawthorne Country Day School, for her help in preparing this manuscript.
H. Shimizu, C.S. McDonough / Research in Developmental Disabilities 27 (2006) 175–189
189
References Alcorn, C. L., & Nicholson, C. L. (1972). Validity of the Slosson Drawing Coordination Test with adolescents of below-average ability. Perceptual and Motor Skills, 34, 261–262. Dube, W. V., Iennaco, F. M., Rocco, F. J., Kledaras, J. B., & McIlvane, W. J. (1992). Microcomputer-based programmed instruction in identity matching to sample for persons with severe disabilities. Journal of Behavioral Education, 2, 29–51. Dube, W. V., McDonald, S. J., & McIlvane, W. J. (1991). Constructed-response matching to sample and spelling instruction. Journal of Applied Behavior Analysis, 24, 305–317. Durfee, J. L., & Billingsley, F. F. (1998). A comparison of two computer input devices for uppercase letter matching. The American Journal of Occupational Therapy, 53, 214–220. Joiner, R., Messer, D., Light, P., & Littleton, K. (1998). It is best to point for young children: A comparison of children’s pointing and dragging. Computers in Human Behavior, 14, 513–529. King, J. (1992). Preschooler’s use of microcomputers and input devices. Journal of Educational Computing Research, 8, 451–468. Lancioni, G. E., & Boelens, H. (1996). Teaching students with mental retardation and other disabilities to make simple drawings through a computer system and special catds. Perceptual and Motor Skills, 83, 401–402. Lane, S. D., & Critchfield, T. S. (1998). Classification of vowels and consonants by individuals with moderate mental retardation: Development of arbitrary relation via match-to-sample training with compound stimuli. Journal of Applied Behavior Analysis, 31, 21–41. Leung, J. (1994). Teaching simple addition to children with mental retardation using a microcomputer. Journal of Behavioral Education, 4, 355–367. MacKenzie, I. S. (1992). Fitts’ Law as a research and design tool in human-computer interaction. HumanComputer Interaction, 7, 91–139. Mastropieri, M. A., Scruggs, T. E., & Shiah, R. (1997). Can computers teach problem-solving strategies to students with mild mental retardation? Remedial and Special Education, 18, 157–165. Merzenich, M. M., Jenkins, W. M., Johnston, P., Schreiner, C., Miller, S. L., & Tallal, P. (1996). Temporal processing deficits of language-learning impaired children ameliorated by training. Science, 271, 77–81. Missiuna, C. (1994). Motor skill acquisition in children with developmental coordination disorder. Adapted Physical Activity Quarterly, 11, 214–235. Neef, N. A., Bicard, D. F., & Endo, S. (2001). Assessment of impulsivity and the development of self-control in students with attention deficit hyperactivity disorder. Journal of Applied Behavior Analysis, 34, 397–408. Saunders, K. J., Johnston, M. D., Tompkins, B. F., Dutcher, D. L., & Williams, D. C. (1997). Generalized identity matching of two-dimensional forms by individuals with moderate to profound mental retardation. American Journal on Mental Retardation, 102, 285–291. Shimizu, H., & Yamamoto, J. (2000, May). Developing a computer program for teaching students with developmental disabilities to use a mouse. Poster session presented at the 26th annual convention of the Association for Behavior Analysis, Washington, DC. Stromer, R., Mackay, H. A., Howell, S. R., & McVay, A. A. (1996). Teaching computer-based spelling to individuals with developmental and hearing disabilities: Transfer of stimulus control to writing tasks. Journal of Applied Behavior Analysis, 29, 25–42. Stronmen, E. F., Revelle, G. L., Medoff, L. M., & Razavi, S. (1996). Slow and steady wins the race? Three-yearold children and pointing device use. Behavior and Information Technology, 15, 57–64. Tallal, P., Miller, S. L., Bedi, G., Byma, G., Wang, X., Nagarajan, S. S., et al. (1996). Language comprehension in language-learning impaired children improved with acoustically modified speech. Science, 271, 81–84. Touchette, P. E., & Howard, J. S. (1984). Errorless learning: Reinforcement contingencies and stimulus control transfer in delayed prompting. Journal of Applied Behavior Analysis, 17, 175–188. Vargas, E. A., & Vargas, J. S. (1992). Programmed instruction and teaching machines. In R. P. West & L. A. Hamerlynck (Eds.), Designs for excellence in education: The legacy of B.F. Skinner (pp. 33–69). CO: Sopris West.