Ship trajectory for a single paddle shipwreck game. 

Context in source publication

Context 1

... how to execute that movement. Performance during this stage is likely to be highly variable with a large number of errors. At the associative stage, skills become more refined with practice, resulting in greater consistency of performance and fewer errors. At the autonomous stage, the motor skill has been learned and little cognitive effort is required to execute it. Krebs and colleagues described these stages of learning as variations of implicit and explicit learning [16]. 2) Types of tasks: Schmidt and Lee [17] classified several types of tasks that can affect how a skill is learned. Discrete tasks have a recognizable beginning and end (e.g., kicking a ball). In contrast, continuous tasks do not have an inherent start and finish as part of the task (e.g., playing the violin). Serial tasks are a collection of discrete tasks that have a particular order (e.g., dressing). Tasks can also be classified as open versus closed, depending on predictability of the environment. Open tasks take place in a constantly changing environment. One cannot plan an entire movement in advance but must rapidly adapt his/her plan in response to the changing environment (e.g., playing tennis). Closed tasks are in a stable environment, which offers predictability to the movement pattern (e.g., free throws in basketball). 3) Practice: Practice schedules are a well-studied area of motor learning. Blocked practice involves repetitive practice on the same task. It results in improved acquisition of motor performance in a short period of time but does not necessarily promote retention [18]. Greater retention and transfer of skills are accomplished with random practice, which involves varying the task demands over practice trial [19]. The effects of blocked versus random practice for children is less clear; some studies have found no difference between these practice schedules for children [20], [21], whereas others have found similar results as in adults, with random practice facilitating greater motor learning [22], [23]. Evidence suggests that the different results may be related to the complexity of the task and the age of the child [24], [25]. 4) Feedback: Intrinsic feedback is information provided by the sensory systems as a result of movement [26]. Extrinsic feedback supplements intrinsic feedback and forms the basis for explicit learning [26], which is learning that results from clearly stated directions or instructions (e.g., verbal feedback). Two sources of extrinsic feedback used in motor learning studies are knowledge of results [27] and knowledge of performance [28]. Knowledge of results is feedback about the final outcome of movement in terms of the environmental goal. In contrast, knowledge of performance is feedback related to the nature of the movement produced [17]. Feedback can be given during the movement (concurrent), right after the movement (immediate), at the completion of movement (terminal), or after a delay [17]. Feedback can also be given consistently (i.e., after every trial) or sporadically (i.e., after some but not all trials). Sporadic feedback after a delay is in general superior for motor learning to consistent feedback given immediately after the movement [29], [30]. The delay in feedback given over some trials allows the individual to determine what factors are influencing performance and prevents reliance on external feedback to learn the skill. While sporadic feedback is superior for adult motor learning, recent evidence suggests that children may respond differently [31] and require a more frequent pattern of feedback. Three goal-directed SGs were designed to address motor impairments including poor coordination, impaired motor speed or accuracy, and diminished strength, as well as to address cognitive or perceptual impairments. Games were developed in TCL/TK. For each of the games, the child moves a paddle using his/her ankle. Dorsi-plantar flexion and inversion-eversion control screen movements of the paddle in vertical and horizontal directions, respectively (see Fig. 1). Although not further discussed in this paper, sporadic feedback about performance was provided during the therapy session. Presentation of the three games follows. 1) The Shipwreck: In this game, the child has to prevent a ship from crashing into the rocks, by using a set of paddles (ship barriers) (see Fig. 2). Dorsi-plantar flexion (inversion- eversion) control simultaneously the vertically (horizontally) moving red and green (yellow and blue) barriers. If the ship is saved by a barrier or falls on the rocks, 20 points are added or subtracted on the scoreboard, respectively. After deflected by a paddle or hitting a rock, the boat bounces with random coordinates ( α , β ). To reduce cognitive load and facilitate motor planning, the game allows play with less than four barriers. In that case, the rocks on an “unprotected” side are replaced by a sandy beach (not shown) that deflects the boat towards the opposite direction. In Fig. 3, the schematic representation of a boat trajectory (blue line) in a single barrier game is shown. After the boat hits the barrier, a new set of ( α , β ) are used to calculate the final position of the boat (that is the new target for the game). 2) Race for Noah’s Ark: In this game, the child runs through a race to collect animals followed by water splashes (see Fig. 4). The paddle represents a racer that is collecting the animals and can move either horizontally or vertically, during the game. The scope of the game is to collect the animals but avoid the splashes that follow it. Animals come in predefined order (up-down-up, etc.) or randomly. Collecting an animal rewards the player by adding 20 points; failing to collect an animal or falling into the splash subtracts 20 or 10 points, respectively. 3) Soccer 2014: In this game, the child plays a soccer game with the computer as his/her opponent (Fig. 5). The child must play both defense and attack. During the defense phase, the player controls the goalkeeper which has the shape of an elliptical paddle (see Fig. 5). The opponent advances towards the center of the field and selects randomly to send the ball to the left or right corner of the goalpost. The goalkeeper now has to move left or right (by inversion/eversion of the ankle) to defend the ball. After the ball is defended, the child has to move his/her ankle near the neutral position so that the goalkeeper will be approximately at the center of the goal. When positioned near the center, the goalie becomes an attacker and takes a circular shape (not shown). The attacker has to kick the ball (dorsiflexion of the ankle) towards the opponent’s goalpost. The kick is visualized as the circular attacker moves upwards until it crosses the middle-line. The ball is released with a speed analogous to the speed of the actual kick. Since the opponent goalkeeper is moving at a constant speed, the probability of scoring increases with the speed of the kick. If the computer goalie is able to defend the shot, the child needs to plantarflex the ankle and move the attacker to the neutral position to defend again his/her goalpost. The three SGs were designed to offer the youngster the opportunity to exercise while challenging his/her ability to move faster or more accurate. Depending on one’s ability to aim, the target of the game may become smaller or larger. Depending on one’s ability to move fast er, the speed of the game may also change accordingly (for further details see [32]). III. D ISCUSSION The SGs presented above are meant to serve as a toolbox for habilitation of children with CP. The present games require discrete movements and can be assisted-as-needed by the robot. In addition, the task objectives of the race, the soccer and the shipwreck games were blocked, serial, and random, respectively, covering the entire spectrum of the structured practice. The predictability of the available gameplay environments also varied greatly. The race game could provide a closed environment whereas the other two game environments were open, with that of the Shipwreck varying the most. While we have not yet clinically tested the pediatric anklebot, one could use the blocked practice race game for the first (cognitive) stage of motor learning, introduce the soccer game during the associative phase and keep the last and most varying game for the autonomous stage. Our proposed goal (aim) for the SGs is that they be met by children as challenges to be mastered rather than threats to be avoided. Unlike other rehabilitation SGs that used 3D engines [33]-[36], we selected the game interface to be a simple 2D screen. Complex graphics are not the only option to attract the attention of children, especially in the ages of 5 to 8. Difficulty level modifications are used in other motor- rehabilitation approaches to ensure patient engagement [33], [34], [36], [37]. But trying to maximize the engagement (assumed to expedite habilitation in children) requires an additional perspective. For the set of proposed SGs, meaningful play and challenge concepts were employed to maximize engagement and sustain attention. Meaningful play employed feedback cues given by both auditory and visual means — buzzers with different sounds for success and failure, paddles changing shape to indicate interaction (in soccer), rocks shaking after being hit by the ship, and scoreboards providing knowledge of results. To maintain an appropriate level of challenge, game elements were changed dynamically according to the ability of the youngster to move and point with the ankle. We modified the size of the paddles (and as a consequence, of the target) and the speed of the game defined as the speed of falling gates in the race game or the speed of the boat or the ball in the other games. Knowledge of performance was also employed to challenge children to improve their performance or, at the very least, maintain it (reducing any slacking) [32]. Two of the ...