![connect word game connect word game](http://a3.mzstatic.com/us/r30/Purple127/v4/2e/13/d6/2e13d65b-f76a-9c05-50f8-2084adaf5f4b/screen696x696.jpeg)
We then normalize the observed average memory buffer size for an individual player to obtain a metric of memory. Using this approach to the analysis, it is possible to determine the maximum and average buffer sizes required for perfect performance. When the matching cards are removed from the board, the corresponding cards are dropped from the buffer. For example, a card that has been seen at a given location would enter the buffer and remain there until its match would be uncovered. The derivation of this measure is based on the concept of a hypothetical memory buffer for an ideal player. The average “life expectancy” of the items in the memory buffer derived from a survival function is used to characterize the working memory size. The model of the cognitive performance on this game is based on a “leaky” memory buffer. These range from simple shape and color matches to cognitively more difficult matches, such as matching a digital clock time with the analogue picture equivalent. Game difficulty is adapted based on number of cards and the cognitive difficulty of the matches. Users must remember the location of various cards they select (turn over to view the face of the card) and then match pairs. For this game, we adapted the standard card game of Concentration, as shown in Figure 4. Although many of the cognitive computer games and standard tests involve short-term and working memory we rely on one game in particular to provide us with a more direct measure of working memory. The time difference in progressing to the task of set switching between numbers and letters is indicative of the relative cognitive difficulty of tracking two simultaneous sequences. However, adding distracters to the task allows us to measure the effect of visual search without changing the memory requirement. Following the sequence of numbers requires a combination of working memory and visual search. For example, measuring an individual subject’s speed in following a single highlighted target with the mouse device provides a baseline measure of motor speed. In addition, we can measure the incremental effects of changing the complexity of the task. In our game environment, we are able to measure each move and model the performance dependent on the complexity of the search. Standard test scores only reflect overall timing and number of errors for this test – a two dimensional representation of the complex processes. This process of set switching (from numbers to letters) requires memory, visual search, motor speed, divided attention, and mental flexibility.
![connect word game connect word game](http://cdn.ghstatic.com/images/screens/300269/1/3.jpg)
![connect word game connect word game](http://static-3.app4smart.me/uploads/posts/thumbs/9625/f-8308-f34b85ce3e.jpg)
The Trail Making Test requires subjects to connect a sequence of numbered circles as quickly as possible, and then connect a sequence of alternating numbers and letters (e.g., 1, A, 2, B, 3, C.). As an example of a game that is closely related to one of the neuropsychological tests, and at the same time measures a combination of multiple processes, is our adaptation of the standard Trail Making Test into a game (shown in Figure 3). Most of the computer games we have created require multiple cognitive processes and yield multiple measures of cognitive performance, reflecting the nature of everyday tasks. In addition to the word-related metrics, we examine the dependency of response time on the search complexity as an indicator of search and planning cognitive abilities. This allows us to focus the user’s attention and narrow the search field. Longer words receive a higher score in addition, subjects are also challenged to use particular letters (green letters for bonus points and red letters to avoid losing the game). Figure 2 shows a slightly more complicated word game, where the user is asked to create words by selecting adjacent letters in sequence. Our complexity measure is related to the entropy and orthographic complexity of the words generated by the user, and defined as h ( w ) = − ⎡ ⎣ log p ( w ) + log q ( w ) ⎤ ⎦, where p corresponds to the word frequency and q to the frequency of the bigrams within the word in the English language (greater rare word usage corresponding to higher cognitive function). However, we also measure the word complexity of the generated words. The simple measure of rate of word generation in this game most directly corresponds to standard measures of verbal fluency. Figure 1 shows a word jumble game, where the user is challenged to create as many words a possible from a scrambled set of 7 letters. Two of our computer games have fairly direct measures of verbal fluency (ability to generate a class of words within time constraints). metrics of associated with each game are based on computational models of basic cognitive processes associated with the neuropsychological tests.