NZ735944B2 - System and process for cognitive assessment and training - Google Patents

Abstract

process for cognitive assessment and training, the process being executed by at least one processor of a computing system and including the steps of: receiving interaction data representing interactions between an application executing on an electronic device and an individual interacting with the executing application; processing the interaction data to generate performance data representing quantitative measures of the performance of the individual with respect to the executing application; and processing the performance data for the individual to generate cognitive assessment data indicative of at least one attention-related ability of the individual. executing application; processing the interaction data to generate performance data representing quantitative measures of the performance of the individual with respect to the executing application; and processing the performance data for the individual to generate cognitive assessment data indicative of at least one attention-related ability of the individual.

Description

SYSTEM AND PROCESS FOR COGNITIVE ASSESSMENT AND TRAINING Technical Field The present invention relates to a system and s for assessing cognitive performance of individuals and for training individuals to improve their cognitive performance.

Background The reference to any prior art in this specification is not, and should not be taken as, an acknowledgment or any form of suggestion that the prior art forms part of the common general knowledge in Australia.

The diagnosis and treatment of developmental lities is an ant problem faced by modern society. The diagnosis of developmental disabilities has increased significantly over the past decade, and executive function weaknesses, such as ion difficulties are a particularly common e characterising the ive impairments of many affected individuals.

There is now widespread ent that there are three core cognitive attentional processes that may be impaired in those with developmental disabilities, namely: i) selective attention, which determines the ability to selectively attend to s of the environment; ii) sustained attention, which enables the individual to focus on a task and to remain sensitive to ng information; and iii) executive attention, relating to the ability to focus on a fixed goal while ignoring conflicting information. ulties in any one of these attentional processes in childhood have been shown to have detrimental effects on learning and social outcomes during school years and beyond. The degree of attention deficiency experienced by an dual will depend on the extent to which they are affected by a developmental disability, and on the presence of any other intellectual disabilities. Attention deficits are highly prevalent in a range of developmental disorders, including Autism Spectrum Disorders (ASDs), Down syndrome, Williams syndrome, and Fragile X syndrome.

The early diagnosis and treatment of attention deficiency is essential for several reasons. First, it may lead to improved educational and social opportunities, and therefore to a better quality of life, for individuals affected by developmental disabilities. There are several difficulties with current approaches to assessing developmental lities. In particular, there is a general lack of objective methods for assessment, which typically involves the subjective assessment of an dual’s state of affliction by a medical health professional. These assessments are difficult to repeat frequently, and each assessment requires a consultation with the medical professional. gh these behavioural ratings are informative, alone they are not sufficient because ion ulties can stem from a number of underlying cognitive weaknesses. For instance several children with developmental disabilities share common profiles of inattention and hyperactivity, yet syndrome specific cognitive attention profiles have been shown. Therefore relying on purely behavioural ratings may result in overlooking core cognitive difficulties.

Traditionally, treatment for developmental disabilities has been implemented in the form of pharmaceutical intervention. r, this type of ent has the disadvantage of being limited in its ability to accommodate for differences in ion impairment for individuals who suffer from other intellectual disabilities, and may be able for patients who have adverse ons to the medications.

Pharmaceutical interventions also only target behavioural weaknesses, and although stimulant medication has been shown to be effective in typically developing children in the short term, the long term effects of this intervention are not known.

The traditional approaches to diagnosing and treating developmental disabilities of an individual in isolation are also problematic. These methods involving case-bycase assessments are uate to effectively deal with the growing issue of childhood developmental disability at a national or global scale.

It is desired to provide a system and process that alleviate one or more difficulties of the prior art, or to at least provide a useful alternative.

Summary In accordance with some embodiments of the present invention, there is provided a process for cognitive assessment and training, the process being executed by at least one processor of a computing system, and including the steps of: receiving interaction data enting interactions n an application executing on an electronic device and an individual interacting with the ing application; processing the interaction data to generate mance data representing quantitative measures of the performance of the individual with respect to the ing application; and processing the performance data for the individual to te cognitive assessment data indicative of at least one attention-related ability of the individual.

In some embodiments, the application is a game, and the interaction data represents ctions between the game and the individual playing the game, the game being configured to assess attention-related abilities of the individual.

In some embodiments, the quantitative measures of the performance of the individual with respect to the executing application e quantitative measures of cy, error rate, and response time.

In some embodiments, the step of processing the performance data includes performing multivariate analysis of the quantitative measures to generate the cognitive assessment data.

In some embodiments, the ariate analysis includes at least one of a principal component analysis and a clustering process.

In some ments, the step of processing the performance data includes processing the performance data for the individual and corresponding performance data for one or more other duals having one or more cognitive ability classifications, including a neurotypical classification and/or one or more developmental lity classifications, the cognitive assessment data being indicative of a classification of the individual with respect to the one or more cognitive ability classifications.

In some embodiments, the process includes generating display data representing a visualisation of the cognitive assessment data of the individual and the one or more cognitive ability classifications of the one or more other duals.

In some ments, the process includes generating display data representing a isation of one or more of the quantitative measures of performance of the individual and one or more corresponding quantitative measures of performance for one or more other duals to allow a user viewing the visualisation to compare the performance of the individual to the corresponding performance of the other individuals.

In some embodiments, the isation includes an interactive control for selecting the one or more quantitative measures of performance for display to the user.

In some embodiments, the one or more other individuals have one or more cognitive ability classifications selected by a user from a set of cognitive ability classifications.

In some embodiments, the visualisation is ured to visually differentiate any quantitative measures of performance of the individual that differ significantly from the corresponding quantitative measures of mance for the other individuals.

In accordance with some embodiments of the present invention, there is ed a s for assessing and training cognitive mance of an individual, the process being executed by at least one processor of a computing system and including: displaying a plurality of visual stimuli on a display of the computing system; receiving inputs of an individual using the computing system, the inputs being responsive to the displayed visual stimuli; generating interaction data representing the visual stimuli and the corresponding inputs of the individual; and sending the ction data to a data processing system configured to process the interaction data to generate performance data representing quantitative measures of the performance of the individual with respect to the visual stimuli.

The computing system may be a tablet computing device (e.g., an iPad) or a smartphone.

In ance with some embodiments of the present invention, there is provided a process for assessing and training cognitive performance of an individual, the s being executed by at least one sor of a computing system and including: displaying a plurality of visual stimuli on a display of the computing system; receiving inputs of an individual using the computing system responsive to the displayed visual i; ting ction data representing the visual stimuli and the corresponding inputs of the dual; and processing the interaction data to generate performance data representing quantitative measures of the performance of the dual with respect to the visual i.

In some embodiments, the visual stimuli represent a game being played by the individual, the visual i being ured for assessing and training attentionrelated abilities of the individual.

In accordance with some embodiments of the present invention, there is provided a computer program product for cognitive assessment and training of an individual, including executable instructions that, when executed by at least one processor of a computing system, performs any one of the above processes.

In accordance with some embodiments of the present invention, there is provided a cognitive assessment and training system, including: a random access memory; at least one processor; a display to display application content to a user of the system; at least one input device to receive input from the individual; wherein the system is configured to execute any one of the above processes.

In some embodiments, the system is a tablet computer and the display and input device are components of a touchscreen of the tablet computer.

In accordance with some embodiments of the present invention, there is provided a system for cognitive assessment and training, including: a data receiving component configured to receive interaction data representing interactions between an application executing on an electronic device and an individual interacting with the executing application; a statistical processing component configured to process the interaction data to generate performance data representing quantitative measures of the performance of the individual with respect to the executing ation; and a classification component configured to s the performance data for the individual to generate classification data indicative of at least one developmental disability classification for the individual.

In accordance with some ments of the present ion, there is provided a method for cognitive assessment and training of an individual, including: providing cognitive training sessions in which the individual continuously interacts with a cognitive assessment and training system for at least a predetermined period of time, the ive assessment and training system being ured to execute any one of the above processes, wherein the executing process implements a computer game being played by the individual, and the computer game is configured to train attention-related ies of the individual playing the game; and the step of processing the ction data is performed at least before and after the ive training sessions to assess improvements in one or more of the attention-related abilities of the individual.

Brief Description of the Drawings Some ments of the present invention are hereafter described, by way of non-limiting example only, with reference to the accompanying drawing in which: Figure 1 is a schematic diagram of a cognitive assessment and ng system in accordance with an embodiment of the present ion; Figure 2 is a block diagram of a computer system used to implement the user device, interaction device and/or the server devices of the cognitive assessment and training system in the described embodiments; Figure 3 is a schematic diagram of the functional components of the ive assessment and training ; Figure 4 is a flow diagram of a cognitive assessment process executed by the cognitive assessment system in accordance with an embodiment of the present invention; Figure 5 is a schematic diagram of a data storage and management component of the cognitive assessment and training system; Figure 6 is a flow diagram of an authorisation process by which users register with and are authenticated to access the cognitive assessment and training system; Figure 7 is a screenshot of a user login screen of the cognitive assessment and training system; Figure 8 is a flow m of a game installation and registration process executed by the ive assessment and training system; Figure 9 is a flow diagram of a process of the cognitive assessment process of Figure 4 for generating interaction data; Figures 10A to 10D are screenshots of game applications of the cognitive assessment and training system; Figure 11 is a flow diagram of a process of the cognitive assessment process of Figure 4 for generating performance data representing quantitative measures of the performance of an individual with t to a cognitive assessment and training application; Figure 12 is a flow diagram of a process of the cognitive ment process of Figure 4 for generating classification data indicative of at least one cognitive fication for the individual; Figure 13a is a screenshot of a user interaction playback y of the cognitive ment and training system, showing the sequence of individual actions of an individual interacting with a cognitive assessment and training application; Figure 13b is a screenshot of an interaction statistics display generated by the cognitive assessment and training system, showing the performance of the dual as a function of time during interaction with a cognitive ment and ng application; Figure 14 is a screenshot of a user analysis display generated by the cognitive assessment and training system, showing the interaction performance of an individual over a series of sessions with a selected cognitive ment and training application, and highlighting statistically significant deviations from the interaction performance of a corresponding selected reference population of users; Figure 15 is a flow diagram of an ction performance visualisation process of the cognitive assessment and training system; Figure 16 is a screenshot of a display ted by the cognitive assessment and training system, showing clustering of the interaction performance of individuals with corresponding developmental difficulties with respect to multiple interaction performance parameters measured by the cognitive ment system; Figure 17 is a flow diagram of a report generation process of the cognitive assessment process of Figure 4; Figure 18 is a screenshot of an example report display generated by the report generation process of Figure 17, showing an dual’s eight performance indicators in the form of a spider plot, and the improvement in the dual’s performance over several sessions of gameplay using the cognitive assessment system; Figure 19 is a partial screenshot of a graphical representation of touch events as a function of time during ay, representing the individual’s touchscreen inputs with labels indicating classifications of those inputs; Figure 20 is a partial screenshot of a graphical representation of the total number of d touch events made by the individual during a game for successive gameplays by the individual, showing a reduction in the number of ‘invalids’ (invalid touch ) over time as the individual learns the game and improves their performance; and Figure 21 is a partial screenshot of a graphical representation of the relationship between elected quantitative measures of the performance of an individual (in this example, the measures being ‘hit accuracy’ on the x-axis and ‘invalids’ (invalid touch events) on the y-axis) while playing the Find a Fish game from day 0 to day 25 of a training program, demonstrating a significant reduction in the number of invalids and a corresponding increase in hit accuracy as the dual improves their performance over a series of training sessions.

Detailed Description The described embodiments of the present ion include a cognitive assessment and training system and s that objectively determine quantitative measures of cognitive mance of individuals of both typical and atypical cognitive abilities by non-invasively measuring each individual’s interactions with a cognitive assessment and training application (e.g., a game) executing on an electronic device (typically, a tablet computer), referred to hereinafter for convenience of description as an “interaction device”. An individual’s responses to stimuli provided by the cognitive assessment and training application are captured and processed to generate corresponding quantitative measures of the dual’s ive performance, and to generate classification data indicative of at least one cognitive classification for the individual.

The described system and process provide an assessment of the cognitive mance of the individual based on their ction performance, and can employ various types of is to produce a variety of quantitative mance measures for the individual. The cognitive performance summary of the individual can be stored by the system, and the ment repeated over time in order to provide a record of any changes in the performance of the individual over time, for example in response to treatment. A l practitioner or researcher can view the individual’s stored cognitive performance for the purpose of making a diagnosis, or to track their progress, for example.

The cognitive ment and training system and process can be d to generate ive performance data for a plurality of individuals afflicted by potentially distinct developmental disabilities and/or intellectual disability conditions, allowing comparative analysis between the ive performance of individuals based on their developmental condition or any of a number of other traits, such as age, , and/or length of treatment, for example. Medical practitioners and clinical researchers can e the performance data generated by the system to model cognitive conditions in order to r improve diagnosis and treatment for individuals affected by developmental disorders.

As described below, the cognitive assessment and training system and process described herein provide customisation options that allow clinical users to modify the analysis methods in response to visualisations of the cognitive ability data, for example by eliminating redundant parameters.

Although embodiments of the present invention are lly described herein in the context of individuals interacting with cognitive assessment and training ations in the form of electronic games, other types of cognitive assessment and training applications could be used in other embodiments, providing other forms of stimulus and measuring responses thereto. However, the use of electronic games promotes engagement of the individual to be assessed, and is particularly advantageous in the assessment and treatment of children. As bed below, the cognitive game applications allow quantitative evaluation of specific cognitive functions of the individual. This allows a clinical medical professional to implement a diagnosis and treatment program based on the specific needs of the individual, as indicated by the cognitive performance data generated by the cognitive assessment system and s.

In the described embodiments, the cognitive performance of an individual is assessed with respect to the ion executive function of that individual. That is, the interaction devices execute applications that present situations conducive to measuring an individual’s selective, sustained and/or controlled ion based on the individual’s measured responses to those situations or stimuli. r, it will be apparent to those skilled in the art that the described ses can be y applied to the assessment of other measurable cognitive functions, such as for example socialisation and g memory executive functions.

Although the ments described herein utilise electronic games, it will be apparent to the skilled addressee that other types of electronic applications can be used to obtain responses representative of an individual’s cognitive attention capabilities. For example, measurements of cognitive attention response can be obtained from an individual engaging in a concentration related task wherein the interactive situations or stimuli are not presented in the t of a game environment.

The cognitive assessment and training system and process allow medical tioners, researchers, and carers to quantitatively compare the cognitive functions of an individual to other individuals with similar or related developmental disabilities for the purposes of improving treatment and/or diagnosis, allowing accurate identification of trends in the treatment and severity of the disabilities, and ng improved research outcomes by providing a large tion of clinical data. Additionally, the system and process are also effective at training individuals to improve attention-related aspects of their cognitive abilities, as described below, and these improvements have been shown to remain up to at least 3 months after use of the cognitive ment and ng system and process had ceased.

The cognitive assessment and ng system and s described herein: 1) non-invasively collect interaction data representing an individual’s interactions with an electronic application (e.g., game); 2) process the interaction data to te performance data representing quantitative es of the individual’s performance with respect to the application; 3) process the performance data for the first individual to generate fication data indicative of at least one developmental disability classification for the first individual; this can be used to identify developmental disabilities and related disorders affecting the individual based on a comparison of their assessed performance to the corresponding performance of other individuals with known assessments of developmental disabilities or related disorders, and to evaluate the cognitive performance of the dual over time; 4) generate reports summarising the performance of an individual; and ) Visually display the performance data of selected duals to clinical users to enable modification of the performance determination means.

As shown in Figure 1, a cognitive assessment and training system 100 includes client devices 114, 116, and server components 101. The client devices e at least one interaction device 114 for use by an individual 118 suffering from a developmental disability, and one or more user s 116 for use by at least one user 120, who may be a clinical user such as a medical professional providing diagnosis or ent to the individual 118, or a researcher.

The interaction 114 and user 116 devices communicate with the server components 101 over a ications network 110 such as the Internet. In the described embodiment, the server components 101 include a web server 102, which provides a user ace for system access, an interaction data server 104 which receives interaction data from the interaction device 114, and stores it in an associated data repository 106, an analysis engine 107 which analyse the received interaction data, and a reporting server 108 which generates cognitive performance data and reports based on the analysis performed by the analysis engine 107.

In the described embodiment, each of the server components 101 is a standard er system such as an Intel Architecture IA-32 based computer system 2, as shown in Figure 2, and the processes executed by the system 100 are implemented as programming instructions of one or more software modules stored on non-volatile (e.g., hard disk or solid-state drive) storage 204 associated with the corresponding computer system, as shown in Figure 2. Howeve r, it will be nt that at least some of the steps of any of the described ses could alternatively be implemented, either in part or in its entirety, as one or more dedicated re components, such as gate configuration data for one or more field programmable gate arrays (FPGAs), or as application-specific integrated circuits (ASICs), for example. It will also be apparent to those skilled in the art that in other embodiments the various components of the cognitive assessment system 100 may be distributed or combined in a variety of alternative ways other than those bed herein, and at different locations.

Each er system includes standard computer components, including random access memory (RAM) 206, at least one processor 208, and external interfaces 210, 212, 214, interconnected by at least one bus 216. The al interfaces include a wireless network interface connector (NIC) 212 which ts the system 100 to the communications network 220, and a display r 214, which may be connected to a display device such as an LCD panel display 222, which may be a creen panel y. Depending on the specific type of computer system, the external interfaces may also include universal serial bus (USB) interfaces 210, at least one of which may be connected to a keyboard 218 and a pointing device such as a mouse 619.

Each computer system may also include a number of standard software modules 226 to 230, including an operating system 224 such as Linux or Microsoft Windows.

The web server 102 includes web server software 226 such as Apache, available at http://www.apache.org, and scripting language t 228 such as PHP, available at /www.php.net, or oft ASP. The data repository 106 includes a structured query language (SQL) database, and an SQL interface 230 such as MySQL, available from http://www.mysql.com, which allows data to be stored in and retrieved from the SQL database.

Figure 3 is a block diagram of the functional components of the cognitive assessment system 100. An interaction application (also referred to herein after for convenience as a “game”) 318 is executed on the interaction device 114 for the purpose of obtaining responses of the individual 118 to stimuli or situations. In the described embodiments, the interaction device 114 is a standard tablet, laptop or other portable computing device capable of executing a game application 318, and typically includes a touchscreen display panel to receive inputs from the individual in response to stimuli displayed on the screen of the portable computing device.

The game 318 presents stimuli or ions to the individual in the form of game scenarios generated by the game components or code 320, and receives input in response to those stimuli from the individual 118. An interaction logging component or logger 322 timestamps and logs the stimuli and corresponding responses, and sends the resulting interaction data to the ction data server 104 for storage, as described below.

The user s 116 execute at least one user application 312 to access the server components 101 in order to perform analyses of the interaction data stored in the data repository and to generate corresponding s for individuals assessed by the system 100. In the described embodiment, the user ation 312 is a standard web browser application such as Google Chrome or oft Internet Explorer. However, in other embodiments the user application 312 may be, for example, a dedicated application that allows the exchange of data between the user s 116 and the server components 101 over a secure communications channel, and is able to display information received from the server components 101 to the clinical user or researcher 120.

The web server 302 provides a single point of entry for a remote user 120 to perform functions including: i) orming and analysing the interaction data received from the interaction device 114 to produce a quantitative assessment of cognitive mance via the is engine 304; ii) storing and retrieving cognitive performance data in and from the data repository 106; and iii) outputting the determined cognitive performance measures in the form of a report via the reporting module 306.

The data repository 106 stores analysis data, including models or representations of the ive performance of each assessed individual, and of general cognitive disability conditions recognised by the system. The R&D (“research and development”) module 310 allows clinical users 120 to create models for new cognitive disability conditions and to modify ng condition models based on the data collected from an assessed individual 118 with an a priori diagnosis, as described below.

For example, when ing the mance metrics of at least one individual to one or more reference data sets for other individuals having respective cognitive classifications, each reference data set can be selected and customised to select aspects of st or to exclude extraneous data; for example, a reference data set can be selected to include performance metrics for all 4 year old individuals assessed by the system 100, all male individuals, or any combination of these or other teristics. The clinical user or researcher 120 can then select a number of data variables to use for comparison between the individual to be assessed and the selected reference data. Although the clinical user or researcher 120 can build these models to look for new cognitive conditions or classifications, a particularly useful feature of the system 100 is to e a quantitative measure indicative of where an ed individual sits on the ASD um.

Reporting data is stored within a report database 510, including the parameters of the ed individual’s performance model over selectable time periods, and clinical notes applied by the practitioner. In the described embodiment, the report storage means 510 is linked to the R&D module 310, allowing the clinical users to modify the format and content of the report based on developments in the field.

Figure 4 is a flow diagram of a cognitive assessment process 400 executed by the server components 101. Configuration steps 401 are med to register individuals to be assessed 118, clinical users 120, and game application information. An individual to be assessed interacts with the game application 318 (by playing the game), and the interaction logger 322 sends the resulting interaction data to the interaction data server 104, which stores the ed interaction data in an interaction data table 506 of the data repository 106.

At step 404, the analysis engine 304 processes the received interaction data to generate performance data representing statistics on various quantitative measures or s of the individual’s performance during their interaction with the game application 318. In the described embodiments, these metrics are ed from a set of metrics that typically includes measures of accuracy, error rate, response time, response erraticness (defined as the average angle between lines joining successive input (e.g., touch) locations), the total number of inputs (e.g., touches), the total game time, the number of game levels played, the highest game level achieved, the number of level attempts, and game progress red as the difference between the ng level and the finishing level of a game played during a session); however, it will be apparent to those skilled in the art that other performance metrics may be used in other embodiments, either in addition to, or instead of, any of these metrics.

At step 406, the mance data is processed to generate classification data tive of at least one developmental disability classification for the individual.

At step 412, the system 100 reports the determined performance data of an individual 412 to one or more users via the reporting module 306. The performance determination 406 and reporting 412 steps can be automatically scheduled by the system 100 such that an dual’s progress is tracked in association with a treatment m or otherwise. In any case, a clinical user 120 with appropriate authority can request an update of the cognitive mance of an dual and/or the generation of a mance report at any time.

Optionally, a clinical user 120 can cause the is engine 304 to generate display data to allow the clinical user 120 to visualise the generated cognitive performance parameters for one or more assessed individuals 408 on a display associated with the user device 116. Based on the isation, at step 410 the clinical user 120 can optionally modify the analysis process by which the analysis engine 304 determines the cognitive performance or classifications of assessed individuals. As described below, this general process can be iterative, allowing a clinician 120 to repeatedly display, filter, transform and modify parameters that influence the cognitive performance or classification(s) of one or more individuals as determined by the system.

In order to use the cognitive assessment system 100, a user first registers with the system 100. Figure 6 illustrates a user registration process 600 by which a new user 601 becomes recognised as a registered user 604 within the system 100 following account registration 606. Account registration 602 involves the new user 601 choosing a user name and/or password combination which becomes associated with that user 601 for future logins to the system 100. Different types of user are ised by the system 100, including duals to be assessed 118 and clinicians who may be further categorised based on their role and level of access to the system data. To log into the system 100, a registered user 604 enters their username and password into xes of a login , as shown in Figure 7, for authentication by the system 100 at step 606, involving a verification of user identifiable information (such as username and password) against the recorded details associated with the user in the user database 504 of the data repository 106.

As shown in Figure 8, configuration of the cognitive assessment system 401 includes the installation of a game application 318 of the system 100 on an interaction device 114 at step 801. This is typically achieved by copying the game application 318 from physical media, such as a CD, DVD-ROM, or removable storage device (for e a USB key) to create a local copy of the game application 318 on the interaction device 114. Alternatively, the interaction device 114 may obtain the game application 318 via communication with an external game server (not shown) of the system 100 over a communications network, which may be the communications network 110 shown in Figure 1.

Once installed, the game application 318 can be executed on the interaction device 114 in the usual way. For example, if the interaction device 114 is an Apple iPad, then at step 901 the game application 318 is ed by tapping a graphical icon representing the game application 318 on the touchscreen display of the interaction device 114.

An individual to be assessed 118 interacts with the game application 318 by simply playing the game implemented by the game ation 318 on the interaction device 114. During gameplay, the game ation 318 presents the dual 118 with a sequence of game ions or stimuli, each prompting a response from the individual. The interaction data is generated by the interaction logger 322 logging each stimulus ted to the individual at step 902, and the corresponding se of the individual at step 904, as shown in Figure 9, until game termination at step 908.

In the described embodiment, these events are logged by including a corresponding logging instructions in the high-level programming language ctions of each game. However, it will be apparent to those skilled in the art that these g events may alternatively be included in a library that includes subroutines or functions referenced by the high-level programming instructions of the game, which can be used to convert games that were not specifically programmed for use with the system 100 two nevertheless be used with the system 100 as described herein.

The resulting interaction data is transmitted to the interaction data server 104 via the communications network 110, and the interaction data server 104 stores the received interaction data in an assessment table 506 of the data repository 106. In the described embodiment, the interaction logger 322 stores the interaction data locally until game termination, at which point the interaction logger 322 sends all of the interaction data for that game session to the interaction data server 104.

However, in other embodiments of the interaction logger 322 may send the interaction data during gameplay.

The ction data for a game session includes information identifying which game object was touched by the individual, when it was touched, how was touched (e.g., whether the individual’s finger was moved during a touch event, and whether multiple fingers were used), and what was displayed on the screen at that time. In the described embodiment, the interaction data is in an XML , although this need not be the case in other embodiments. An excerpt of an XML interaction data file is shown below. <?xml version="1.0" ng="utf-8"?> <trial time="17/06/2014 11:26:46 AM"> <screensize>(1024, 720)</screensize> <mascot>Mascot_Pirate</mascot> <trackerdata> <level>1.0</level> <action> <time>10.05</time> <touchonposition>(289, 75)</touchonposition> <touchoffposition>(289, 75)</touchoffposition> <touchduration>107</touchduration> <touchtype>TargetFish</touchtype> <fishtype>GoldFish</fishtype> <fishposition>(312, 77)</fishposition> </action> <action> <time>14.65</time> <touchonposition>(485, 287)</touchonposition> <touchoffposition>(485, 287)</touchoffposition> <touchduration>153</touchduration> <touchtype>InvalidTouch</touchtype> ningfish> <fishtype>GoldFish</fishtype> <fishposition>(881, 619)</fishposition> </remainingfish> <levelcomplete>yes</levelcomplete> touchcount>7</totaltouchcount> <fishcorrect>6</fishcorrect> <fishincorrect>1</fishincorrect> </trackerdata> <trackerdata> <level>2.0</level> <action> kerdata> </trial> As will be apparent from the XML excerpt, the interaction data includes, inter alia, data identifying the level numbers being played, and within each level, the screen coordinates of each game object (in this example, a TargetFish ), the screen coordinates of each screen touch event by the individual being assessed (including the start and end points of each touch , the temporal duration of each touch event, and timestamps for the display of each object and the start time of the corresponding touch event. If desired, these individual events can be ised graphically by the system 100, as shown in Figure 19. ission occurs via the communications network 110 using a transport layer protocol such as TCP/IP. To transmit the interaction data, the interaction device 114 may utilise a wireless networking ace operable in accordance with an IEEE 802.11 or ‘WiFi’ wireless communications protocol to relay the data to the data server 104 via a local ss k. Alternatively, the interaction device 114 can be connected to a routing or gateway node of the communications network 110 via a direct physical connection, where data transmission occurs to the network 110 via an Ethernet IEEE 802.3 protocol.

The selection of a game to use in the assessment of a particular individual 118 may be based on factors relevant to the dual’s ion, or the specific developmental disability that the treating medical practitioner wishes to test for. In the described embodiment, the game applications 318 provided with the cognitive ment system 100 include game applications that allow clinicians to test for a variety of developmental disabilities and other intellectual disabilities via the assessment of different types of attention executive functions.

In general, each game application 318 provides a fixed linear hierarchy of successive game levels so that each individual playing a game progresses through the same levels in the same order, the only exception being that each level needs to be ‘passed’ before succeeding to the next higher level in the hierarchy. If a level is not passed, then it needs to be repeated before the individual can progress to the next level.

Some games measure selective attention by challenging the individual 118 to differentiate s based on criteria such as colour and size. A deficiency in this cognitive y is associated with autism spectrum disorders. An example of a game that uses differentiation to measure selective attention is the ‘Find a fish’ game, a screenshot of which is shown in Figure 10a.

In the ‘Find a fish’ game, target fish remain constant throughout all trials and are always orange in colour and medium in size. There are a total of 8 target fish per trial, and the individual 118 is required to find 6 of these 8 fish in order to successfully complete each level. ctors vary in frequency and dimension as the levels progress. There are either: None, Some (4), Many (8) or Lots (16) of distractors, and their numbers vary in the proportion that they are similar to the target (0%, 25%, 50%, 75% & 100%). The first dimension that the distractors vary on is colour. The second dimension that the distractors vary on is size. In later trials they vary on both size and colour.

If the individual being assessed does not press ng for 15 seconds, or if 3 utive errors are made, then bubbles appear at the side of the screen to prompt a response from the dual. If nothing is pressed after 30 seconds, or if 3 consecutive errors are made again at any point throughout the trial, then the avatar’s head pops into view from the side of the screen and holds up a sign showing the correct demonstration (finger ng the target fish).

Attentional control is measured by a different type of game, such as a type of Attention Network Test that measures conflict resolution and resistance to distractor inhibition. Examples are the ‘Feed Elvis’ and ‘Sleepy Elvis’ games shown in s 10b and 10c.

The ‘Feed Elvis’ game requires that the individual 118 determine the direction of a target, and make a selection that resolves a problem. The target is Elvis the elephant (central ). Individuals must orient their attention to Elvis and then respond appropriately based on his orientation. If Elvis’s trunk is pointing to the right, then the child is required to select the right peanut bag. Distractors are elephants that are the same as Elvis, and act as flankers. They appear next to Elvis and increase in frequency from 2 to 4 rs. In addition, they also differ in size from Elvis as well as space. Importantly, the direction that the rs face also varies with the flankers either facing the same direction as Elvis (congruent) or the opposite direction (incongruent). Incongruent trials are deemed to be harder, because the child has to overrule the direction that the majority of the elephants are facing and respond only to the direction that Elvis al target) is facing.

If on any trials, including the practice trials, the child is inactive for 15 seconds then a green arrow comes down and points to Elvis in the middle. If the dual presses other items on the screen other than the bags more than 15 times, then the two bags glow green. If either of these occur twice in a row, then the avatar demonstrates the correct response.

In ‘Sleepy Elvis’ the individual 118 is cted that they have to respond as quickly as possible by pressing a target (Elvis the elephant), and to inhibit a response when a no-go stimulus is presented (lion). This game primarily gets harder by ng the display time of the target and the inter-stimulus interval (ISI), being the time between the display of successive stimuli. In the hardest levels, distractions occur as the lion begins to disguise himself as Elvis. These trials incorporate complex aspects of inhibition, and are closely related to stop signal tasks. Individuals are likely to begin making a response when the lion looks like Elvis, however they have to inhibit this response when the disguise falls off. These is a harder task as a motor response has already begun.

If Elvis is not pressed in the given time limit, then verbal instructions occur voice over states ‘Press Elvis as quickly as you can!’ If the individual still does not press the target, then the avatar demonstrates the correct response. ned attention can be tested by games that assess cognitive ‘focus’. An example is the ‘Treasure hunt’ game, as shown in Figure 10d, where the game requires that the individual respond sporadically to game situations.

In the ure hunt’ game, a re chest is presented and the individual 118 is tasked with tapping gold coins that come from the treasure chest. The game ulty increases by increasing the time that the individual 118 has to wait before a target coin appears, and by increasing the number of times it moves in and out of the chest without stopping. In addition, the time that the coin hangs in the air is also d as the game level increases to ensure that the user 118 is paying attention to the task. If the user 118 misses the coin, then that level is repeated until 6 coins are successfully d.

The interaction data transmitted by the game application 318 includes identifiers that identify the individual to be assessed 118 and the game 318 being played, game situations or stimuli presented to the individual 118 during gameplay, and the individual’s 118 responses to each of these stimuli.

Assessment of an individual’s cognitive mance and classification by the system 100 involves the analysis engine 304 executing a process 404, as shown in Figure 11, for generating mance data representing quantitative measures of the performance of an individual with respect to the game application played by the individual.

In the described embodiment, the generated performance data include statistical measures or metrics of accuracy, error rate and response time. Tables 2-5 list the performance metrics generated for the games of ‘Find a fish’, ‘Treasure hunt’, ‘Feed Elvis’ and ‘Sleepy Elvis’ respectively as described above, and Table 1 lists the mance metrics common to all of these games. These metrics can be generated for sampling windows of varying sizes to e multiple parameters for each metric type, as configured by the clinical user or researcher 120. For example, the response times of the individual 118 to tive stimuli can be measured over groups of N situation-response pairs, and an aggregated response time value can be ted by a statistical analysis of the N sample es.

The is engine 304 stores the generated cognitive parameters 1103 in the data repository 106.

Table 1: Performance metrics common to all games Name Description Pos Acc on Accuracy, closeness to centre of target Errors Number of distractor touches per level Invalids Number of touches on the background per level Hit Time time from the last touch Time/lvl time taken to te a level Hit Acc Hit accuracy, % of valid touches per total touches Attempts No. of level attempts per game (includes retried levels) Levels No. of levels completed per game (excludes retried levels) Table 2: ‘Find a fish’ performance metrics.

Name Description Hit Acc, Color Hit cy, filtered for colour levels Hit Acc, Size Hit accuracy, filtered for size levels Hit Acc, Col/Size Hit accuracy, filtered for colour & size levels Hit Acc, 4 Dist Hit accuracy, filtered for levels with 4 distractors Hit Acc, 8 Dist Hit accuracy, filtered for levels with 8 distractors Hit Acc, 16 Dist Hit accuracy, filtered for levels with 16 distractors Hit Acc, 100% Dist Hit accuracy, filtered for levels with 100% dissimilar distractors Hit Acc, 75% Dist Hit accuracy, filtered for levels with 75% dissimilar distractors Hit Acc, 50% Dist Hit accuracy, filtered for levels with 50% dissimilar distractors Hit Acc, 25% Dist Hit cy, filtered for levels with 25% dissimilar distractors Hit Acc, 0% Dist Hit accuracy, filtered for levels with 0% dissimilar distractors Errs, Color Errors per level, filtered for colour levels Errs, Size Errors per level, filtered for size levels Errs, Col/Size Errors per level, filtered for colour & size levels Errs, 4 Dist Errors per level, filtered for levels with 4 distractors Errs, 8 Dist Errors per level, filtered for levels with 8 distractors Errs, 16 Dist Errors per level, filtered for levels with 16 distractors Errs, 100% Dist Errors per level, filtered for levels with 100% dissimilar distractors Errs, 75% Dist Errors per level, filtered for levels with 75% dissimilar distractors Errs, 50% Dist Errors per level, filtered for levels with 50% ilar distractors Errs, 25% Dist Errors per level, filtered for levels with 25% ilar distractors Errs, 0% Dist Errors per level, filtered for levels with 0% dissimilar distractors Angle Total Angle between touches Table 3: ‘Treasure hunt’ performance metrics.

Name Description Hit Acc, Dstr Lo Hit accuracy, filtered for levels with < 12 distractors Hit Acc, Dstr Hi Hit accuracy, filtered for levels with >= 12 distractors Hit Acc, Dur Long Hit accuracy, ed for levels with a target duration > 7 sec Hit Acc, Dur Short Hit accuracy, filtered for levels with target duration <= 7 sec Hit accuracy, ed for levels with time between targets <= Hit Acc, Time Short sec Hit accuracy, filtered for levels with time n targets > Hit Acc, Time Long sec Errs, Dstr Lo errors per level, filtered for levels with < 12 distractors Errs, Dstr Hi errors per level, filtered for levels with >= 12 distractors errors per level, filtered for levels with a target duration > 7 Errs, Dur Long errors per level, ed for levels with target on <= 7 Errs, Dur Short errors per level, filtered for levels with time between targets Errs, Time Short <= 10 sec errors per level, filtered for levels with time between targets Errs, Time Long > 10 sec Table 4: ‘Feed Elvis’ performance metrics.

Name Description Hit Acc, Left hit accuracy, filtered for levels with left facing targets Hit Acc, Right hit accuracy, filtered for levels with right facing targets Hit Acc, 2 fl hit accuracy, ed for levels with 2 flankers Hit Acc, 4 fl hit accuracy, filtered for levels with 4 flankers Hit Acc, Con hit accuracy, filtered for levels with congruent flankers Hit Acc, Incon hit accuracy, ed for levels with incongruent flankers Hit Acc, size1 hit accuracy, filtered for levels with size 1 flankers Hit Acc, size2 hit accuracy, filtered for levels with size 2 flankers Hit Acc, size3 hit accuracy, filtered for levels with size 3 flankers Hit Acc, space1 hit accuracy, filtered for levels with space 1 flankers Hit Acc, space2 hit accuracy, filtered for levels with space 2 flankers Hit Acc, space3 hit cy, filtered for levels with space 3 flankers Errs, Left errors per level, filtered for levels with left facing targets Errs, Right errors per level, filtered for levels with right facing targets Errs, 2 fl errors per level, filtered for levels with 2 flankers Errs, 4 fl errors per level, filtered for levels with 4 rs Errs, Con errors per level, filtered for levels with congruent flankers Errs, Incon errors per level, filtered for levels with incongruent flankers Errs, size1 errors per level, filtered for levels with size 1 flankers Errs, size2 errors per level, filtered for levels with size 2 flankers Errs, size3 errors per level, filtered for levels with size 3 flankers Errs, space1 errors per level, filtered for levels with space 1 flankers Errs, space2 errors per level, filtered for levels with space 2 flankers Errs, space3 errors per level, filtered for levels with space 3 flankers Table 5: ‘Sleepy Elvis’ performance metrics.

Name Description Hit Acc, Elvis hit accuracy, filtered for levels with Elvis s Hit Acc, Lion hit accuracy, filtered for levels with Lion targets Hit Acc, 3sec hit accuracy, filtered for levels with a 3sec display time Hit Acc, 2sec hit accuracy, filtered for levels with a2sec display time hit accuracy, filtered for levels with an inter stimulus interval of 3000- Hit Acc, Slow 4000ms hit accuracy, filtered for levels with an inter stimulus interval of 1800- Hit Acc, Med 2800ms hit cy, filtered for levels with an inter stimulus interval of Hit Acc, Fast 1600ms Hit Acc, dsg 0 hit accuracy, filtered for levels with no disguise Hit Acc, dsg 1 hit accuracy, filtered for levels with elephant ears costume Hit Acc, dsg 2 hit accuracy, filtered for levels with ears & trunk costume hit accuracy, filtered for levels with ears, trunk, head and back piece Hit Acc, dsg 3 costume Hit Acc, dsg 4 hit accuracy, filtered for levels with the entire nt e Hit Acc, 1 hit accuracy, filtered for levels with the disguise time = 500ms Hit Acc, dsgTim2 hit accuracy, filtered for levels with the disguise time = 750ms Errs, Elvis errors per level, filtered for levels with Elvis targets Errs, Lion errors per level, ed for levels with Lion s Errs, 3sec errors per level, filtered for levels with a 3sec display time Errs, 2sec errors per level, filtered for levels with a 2sec display time errors per level, filtered for levels with an inter stimulus interval of Errs, Slow 3000-4000ms errors per level, filtered for levels with an inter stimulus interval of Errs, Med 1800-2800ms errors per level, filtered for levels with an inter us interval of Errs, Fast 1600ms Errs, dsg 0 errors per level, filtered for levels with no disguise Errs, dsg 1 errors per level, filtered for levels with elephant ears costume Errs, dsg 2 errors per level, filtered for levels with ears & trunk costume errors per level, filtered for levels with ears, trunk, head and back Errs, dsg 3 piece costume Errs, dsg 4 errors per level, filtered for levels with the entire elephant costume Errs, dsgTim1 errors per level, filtered for levels with the disguise time = 500ms Errs, 2 errors per level, filtered for levels with the disguise time = 750ms In the described embodiment, the cognitive ment system 100 stores a new set of performance data for each individual 118, each time that the individual 118 plays a game 318 on the interaction device 114. Assessments of cognitive performance involve the is of these sets of parameter data using an analysis s 1200 to generate representations of performance, as shown in Figure 12.

To assess a selected aspect of cognitive performance of a selected individual at a selected time, the analysis engine 304 obtains the corresponding set of performance data from the data repository 106. For example, the clinical user or researcher 120 might choose to assess cognitive performance based on a specific game type as determined by the individual’s condition, and/or over a selected time interval of cognitive ability measurement (such as any time in the last 6 months).

Using the resulting sets of cognitive ter data, a ed individual analysis process 1202 is applied to assess the cognitive performance of the individual 118.

The individual analysis process 1202 can choose to assess the cognitive performance of the individual in isolation, or to apply one or more data analysis 1206 techniques, ing statistical analysis, regression, and/or ring, to produce an assessment of the dual’s cognitive performance relative to other individuals. For example, where clustering analysis is used as a classification or diagnosis tool, the closeness of the individual to each other r of other individuals in an N-dimensional space of selected performance metrics can be assessed by determining the distance between the N-dimensional vector of mance metric values for that individual and the average N-dimensional vector representing the centroid of the cluster. Where the individuals of a r have a common developmental disability diagnosis, the classification(s) or diagnosis/diagnoses of the assessed individual can be assessed (and expressed mathematically) in terms of these ces, or as a provisional diagnosis where the performance metrics of an individual appear to belong to a cluster of individuals with a common known diagnosis.

The cognitive assessment system 100 allows the clinical user or researcher 120 to customise the methods used to perform the cognitive is process 1202. The clinical user 120 can manually analyse the performance of an individual by viewing a -by-second real-time display 1204 of the individual’s game play. For example, Figure 13a shows a screenshot representing the sequence of actions taken by an individual’s while playing a selected portion of a selective attention game, allowing the clinician to observe the individuals interactions and decisions.

The al user 120 can also choose to simultaneously view all or a subset of multiple performance metrics of the individual, as shown in Figure 13b.

Additionally, the clinical user 120 can t a multidimensional data analysis to view correlations between the different performance metrics for the individual, for the purpose of determining or providing an indication of a possible classification of the individual with respect to one or more pmental disability classifications or diagnoses.

The performance determination step 406 can also e a comparative analysis process 1208 to assess the performance of the given individual 118 in comparison to a selected control group of other individuals, or to a representation of a known condition. The clinical user 120 can choose to use preset criteria 1210, such as game type or condition, to perform the comparative analysis, where the analysis involves the ination of statistical differences n the performance s of the given dual and one or more control sets of performance metrics for other groups of individuals (e.g., including groups of individuals having ent types or s of developmental disabilities, and a group of individuals without any developmental disability). atively, custom criteria can be selected at step 1212 for the comparison, such as the selection of specific performance metrics.

The control group can be varied by the application of filters including development disability condition, gender, and age.

The cognitive assessment system 100 provides the clinical user 120 with a visual display of the comparative analysis, as shown in Figure 14. The user 120 can select to highlight in a selected colour performance metrics that lie within (or conversely outside) 1.5 standard deviations of a reference data set of performance metrics for other individuals assessed by the system 100. The clinical user 120 is thus alerted to extreme differences in the parameter values of the cognitive data for the given individual 118 compared to those of the control set. Other visualisations can be used to assess other aspects of an individual’s performance over time. For example, Figure 20 is a screenshot illustrating the number of invalid touch events made by an individual during the course of gameplay of the Find A Fish game, showing a decrease in error rate to a relatively constant rate as the individual learns the game and thus improves in performance.

In the described embodiments, the cognitive performance of an individual is ented as a statistical model, where the model parameters are determined by the analysis engine 304 and are subsequently stored in the data repository 106 at step 1214. Analysis can also be performed automatically by the analysis engine 304 in accordance with configuration options set by the al user 120 for the given individual 118 when required, according to the predetermined schedule, or when specifically requested by a clinical user 120 with appropriate authority. In practice, it has been found that of all the performance metrics described herein, only about 12 of them are required in order to characterise about 95% of the disability characteristic behaviours of children assessed by the system 100, and are thus sufficient to represent a “model” of a child being ed, and to compare with models of other children as described .

A clinical user or researcher 120 can visualise the ive performance data 408 of one or more selected individuals assessed by the system 100 using the visualisation process 1506 shown in Figure 15. The clinical user 120 is presented with a graphical user interface (GUI) that allows the selection of one or more performance metrics (or ‘cognitive parameters’) 1500 to be visualised. The visualisation illustrates the onships between performance parameters. One or more developmental disability conditions can be selected by the clinical user 120 at step 1502, which causes the visualisation process 1506 to limit the displayed data to data ated with duals affected by the selected condition(s).

Multivariate filtering and analysis methods can be applied at step 1504, as selected by the al user or researcher 102, from a set of available analysis methods, including principal component analysis (PCA) clustering techniques, support vector machines, Bayesian analysis, on trees, and genetic algorithms in order to identify and quantify ations n one or more of the performance s and corresponding teristics or disability classifications of assessed individuals. The al user or researcher 120 can thus develop arbitrary associations between performance metrics that provide the desired ability to discriminate between individuals or developmental disability conditions of interest.

This can also be performed automatically by selecting a population of individuals based on selection criteria entered by the clinical user or researcher 120, and then automatically processing the resulting sets of mance metrics in order to select a subset of these performance metrics that provides the best predictive capability within the selected population, based on known diagnoses and/or other characteristics of the individuals. Once identified, the ed subset of performance s can then be used to assess an unknown individual.

Figure 16 shows an example of the visualisation data 1506 ted from two selected parameters (performance metrics). Statistical parameter distributions are shown for each individual colour-coded according to the diagnosis of the individual (if , assisting the clinician to gauge how effectively the performance s distinguish between different diagnosis conditions such as high functioning autism, low functioning autism, downs syndrome, and neurotypical development. The parameter associations identified by the visualisation process can be stored in the data repository 106, if desired. The customised parameter sets can then be used to modify the analysis process 410 via the creation of new statistical models that e the accuracy of the performance determination s 406.

The cognitive assessment system 100 includes a reporting component 108 that executes a report generation process 1700, as shown in Figure 17, to generate reports 412 summarising the cognitive performance of individuals 118. The report tion 1700 involves a report configuration step 1701 that allows a user to select the type of report produced, the frequency at which reports are automatically generated, and the set of clinical users 120 who will receive the report. Each report is specific to an assessed dual 118, and generation of the report es obtaining the relevant performance data 1702 from the data tory 106. In the described embodiment, performance data summarised within the report includes individual and comparative analysis models. For example, a report can show the mance of the individual 118 according to: i) absolute tical measures of cognitive performance; and ii) a comparative ranking of their general (or parameter specific) performance compared to a control group, such as other duals with the same condition. The reported performance may include the exact cognitive performance metrics determined by the system 100, but may also include other information inferred from those performance metrics.

The report generation s 1704 for an ed user 118 varies based on the recipient of the report. For example, reports generated for the l practitioner of an individual 118 can contain additional details, such as clinical notes, which are omitted from reports generated for the parent or guardian of the assessed dual. Figure 18 shows an example report for a parent of an assessed dual, where the report contains metrics derived from the cognitive performance model of the individual 118. The format and content of the report can be customised by clinical users 120 in order to allow the analysis ed by the system to remain consistent with developments in the field.

Reports can be generated automatically by the system 100 at regular intervals using cognitive performance and/or parameter data between the present time and the time of the last generated report for the same user 120 and assessed individual 118. The time interval of reporting can be specified by a al user 120 for each individual. The periodic generation of reports s the progress of an individual 118 to be tracked against the wider population, and allows the assessment of developmental conditions to be continually improved as more data is made available to the system 100. Additionally, a clinician 120 can request a report for a specific individual 118 whom they have treated, and the system can output an immediate performance report based using the time interval indicated by the ian 120.

In addition to its ability to quantitatively assess individuals for characteristics or ses of developmental disabilities, the system 100 is also an effective tool for training individuals in order to improve their cognitive abilities, in particular deficits in attention-based abilities.

In order to demonstrate the applicability of the system 100 in ing attentionbased abilities of individuals with developmental disabilities, nine children with developmental disabilities (Mage=8 years, 5 months) and their parents took part in a focus group which involved using the program and then providing feedback on their experience. In order to assess the construct validity and sensitivity of the program, 90 lly developing children (Mage= 4 years 4 months, 3 years to 5 years) were recruited. The system 100 acquired interaction data for the participants, and this was used to generate corresponding performance metrics as described above. In addition, two standard es of attention were also applied to the participants, namely Wilding Attention Tasks (“WATT”) and the Kiddie uous Performance Task (“K-CPT”).

Qualitative data demonstrated that children with a developmental lity were able to engage with the game applications described above and tand the task ements. As shown in Table 6 below, correlation coefficients revealed significant positive correlations between rd measures of attention and the performance metrics of the system 100 relating to selective attention tasks, r (85) =.48, p<.001, cognitive flexibility tasks, r (86) = .44, , sustained attention tasks, r (84) = .36, p<.001, and response inhibition tasks, r (84) = .47, p<.001.

A series of hierarchical multiple regression analyses were used to predict performance on each of the system 100 tasks. For each of the game application tasks, the addition of age as a parameter icantly improved the prediction. In combination, the two predictor variables of age and gender explained 40% of the variance in selective attention performance [adjusted R2=.386, F (2, 87) = 28.96, p<.001], 12.3% of the variance in sustained attention performance, [adjusted R2=.103, F (2, 87) = 6.102, p=.003], 31.3% of the variance in cognitive flexibility performance, [adjusted 8, F (2, 87) = 19.846, p<.001] and 11.8% of the ce in response inhibition performance [adjusted R2=.098, F (2, 87) = 5.847, Table 6. Correlations between errors on the system 100 subtests and standard measures of attention Standard Attentional Measures performance metric VISEARCH Single K-CPT VISEARCH Dual generated by the Search Search system 100 ion .480*** .380*** .431*** Vigilance .439*** .360*** .259* Conflict Resolution .524*** .509*** .441*** Response Inhibition .421*** * .456*** To demonstrate the sensitivity of the cognitive assessment and training system and process in detecting age-related changes in attention performance, a one way ANOVA was conducted to assess differences in mance across age ranges.

Significant differences across ages were found for all tasks: selective attention (p=.01); sustained attention (p=.03), conflict resolution ) and se inhibition (p=.01).

A ng program in the form of a double blind randomized controlled trial was ted. 80 children with intellectual disabilities (Mage=8.02, range 4 to 10 years, IQ<75) were randomly assigned to an adaptive attention training program using the system 100 (intervention) or a non-adaptive control program. As described above, the game applications used for the program incorporated selective attention, sustained attention and attentional control tasks. The intervention ran for 5 weeks and consisted of 25 sessions, lasting 20 minutes each. en were assessed on a range of standardised and tailored assessments before the intervention, immediately after the intervention and 3 months after the intervention. Both parent and teacher reports of inattentive behaviour were For the intervention group, repeated measures ANOVAs revealed significant main effects of time for all of the cognitive attention variables: feature search errors, F(2, 74)= 11.09, p=.001; feature search time, F(2, 74)=3.20, p=.05; conjunction search errors, (2, 62)=14.61, p=.001; conjunction search time, F (2,64)=5.59, p=.006; vigilance targets, , F(2, 70)=9.11, ; and vigilance errors, F(2, 70)=4.52, p=.014. Post hoc se comparisons with Bonferroni adjustment revealed that the intervention group made significantly fewer errors on the feature search and conjunction search tasks across the trial. In addition, the time taken to complete the conjunction search task was shown to significantly decrease from T1 to T2 (p=.03). In terms of the vigilance task, improvements were present at T3 when compared to T1 for both targets located (p=.002) and errors made (p=.02).

In contrast, improvements were not as readily observed in the control group, with significant main effects of time only being t for one variable; time taken to complete the ction search task, F (2,68)=6.46, p=.003. Pairwise comparisons revealed that marginally statistically significant improvements were seen from T1 to T2 (p=.05) which were maintained up until T3 (p=.03).

In order to assess treatment s on the magnitude of improvements in ion skills across time, repeated measures ANOVAs were conducted with condition (Intervention or Control) as the n subjects variable and time as the ndent variable. Significant interactions were observed for both feature search errors and conjunction search , indicating that the reduction in errors made over time differed significantly across groups. Post hoc tests between subjects contrast for time showed that at T1 participants in the intervention group made significantly more errors when compared to the control group on both the feature search, F(1, 74)=4.14, p=.05 and conjunction search task, F(1, 69)=3.93, p=.05 . However by T 2 there was no ence across groups on either of the search tasks (p>.05), with the intervention group ng the amount of errors they made. These improvements were maintained up till T3, as no ences across groups were present (p>.05). No other interaction effects of time and group were ed for the additional attention variables.

Paired-samples t tests revealed that children in the intervention group showed significant improvements in mance on complex selective attention tasks immediately after ng, (t (15) = -3.25, p<.01). Although improvements were not observed in other attentional processes immediately after training, improvements in basic selective attention (t (15) = -2.85, p<.05) and sustained attention (t (15) = -2.20, p<.05) were evident at the 3 month follow up. No improvements were observed in the control group on any attention task, either immediately after training or at follow up. Behavioural measures of inattentive and hyperactive behaviour completed by parents and teachers indicated improvements in the intervention group after training, r these improvements did not reach significance.

The ed intervention provided by the system 100 produced improvements in core attentional processes in children with developmental disabilities when compared to the control program. Importantly, these preliminary results emphasise the potential of these ng paradigms, and offer an alternative to pharmaceutical interventions in individuals who are ‘at risk’ or already vulnerable to attention difficulties.

Many modifications will be apparent to those skilled in the art t departing from the scope of the present invention.

Claims (9)

Claims 1. A process for cognitive assessment of an individual, the process being executed by at least one processor of a computing system, and including the steps of:
1. I. receiving interaction data representing interactions between an application executing on an electronic device and an individual interacting with the executing application; wherein the electronic device comprises a touchscreen display panel; and n the application is a game comprising one or more levels, displaying game objects including one or more target objects, one or more ctor s, and a ound; and wherein the interaction data represents interactions between the game and the individual playing the game, said interaction data comprising information regarding each touch event including; a) information identifying which part of the touchscreen display panel was touched by the individual, including the screen coordinates of the touch event, and the start and end points of each touch event; b) information identifying when the touchscreen display panel was d; c) information identifying what was displayed on the touchscreen display panel at the time of the touch event; d) the screen coordinates of each game ; e) timestamps for the display of each game object and the start time of the corresponding touch event; and f) information identifying how the touchscreen display panel was touched during the touch event, including but not d to; i. information identifying the temporal duration of the touch event; and/or ii. information identifying whether the dual's finger(s) moved whilst in contact with the touchscreen display panel; and/or iii. information identifying whether multiple fingers were used in a single touch event; II. sing the interaction data to generate performance data representing quantitative measures of the mance of the dual with respect to the executing application, wherein the quantitative measures of the performance of the individual with respect to theexecuting application include quantitative measures of accuracy, error rate, and response time, including at least; a) Position Accuracy; defined as closeness of the touch to the centre of a target object; b) Errors; defined as the number of touches on distractor objects per level; c) Invalids; defined as the number of touches on the background per level; d) Hit time; defined as the time from the last touch; e) Time Per Level; defined as the time taken to complete a level; f) Hit Accuracy; defined as the % of valid touches per total touches; g) ts; defined as the number of level attempts per game, including retried ; h) Levels; defined as the number of levels completed per game, excluding d levels; III. wherein the step of sing the mance data includes processing the performance data for the individual and corresponding performance data for a plurality of other assessed individuals the plurality having cognitive ability classifications, ing one or more developmental disability classifications, wherein the processing comprises a clustering process, wherein the closeness of the individual to each other cluster of the other assessed individuals in an N-dimensional space of selected performance data is measured in terms of the distance(s) n the N-dimensional vector of performance data values for the individual and the average N- dimensional vector representing the centroid of the cluster of the other assessed individuals, wherein the other assessed individuals of a cluster have a common cognitive ability classification, including a developmental disability diagnosis, allowing for the assessment of the individual, in terms of the distance(s), where the performance data of the individual appear to belong to a cluster of the other assessed individuals with a common cognitive ability fication, the cognitive assessment being indicative of a classification of the individual with respect to the cognitive ability classifications.
2. 2. The process of claim 1, wherein the other assessed individuals have one or more cognitive ability fications, the cognitive ability classifications including one or more developmental disability classifications, selected by a user from a set of ive ability classifications.
3. 3. The process of claims 1 or claim 2, r comprising ting display data enting a visualisation of one or more of the quantitative measures of mance of the individual and the corresponding mance data for the other assessed individuals having cognitive ability classifications, wherein the visualisation is configured to visually differentiate any quantitative measures of performance of the individual that differ significantly from the corresponding quantitative measures of performance for the other assessed individuals.
4. 4. The process of claim 3, further comprising receiving an input representing an analysis process modified by a user based on the visualisation; and in se to the input, customising operations of ters that influence a determination of the cognitive performance or classifications of the other assessed individuals; and performing the customised operations for the performance data for the individual to generate a desired cognitive assessment and training data indicative of at least one attention-related ability of the individual wherein the customized operations t of selecting one or more attention-related ies to limit the displayed data to data associated with the other assessed individuals corresponding to the selected abilities.
5. 5. The process of claim 4, wherein the customising operations of parameters comprises selecting one or more parameters ed from the group consisting of an age ter of the other assessed individuals, and a gender ter of the other assessed individuals.
6. 6. A computer program product for cognitive assessment of an individual, including executable instructions that, when executed by at least one processor of a computing system, performs the process of any one of claims 1 to 5.
7. 7. A cognitive assessment and training system, ing: a random access memory; at least one processor; a display to display application content to a user of the ; at least one input device to receive input from the individual; wherein the display and the input device are components of a touchscreen; wherein the system is configured to execute the process of any one of claims 1 to 5.
8. 8. The cognitive assessment and training system of claim 7, wherein the system is a tablet computer and the display and input device are components of a touchscreen of the tablet computer.
9. 9. A method for cognitive assessment and training of an individual, ing: providing cognitive training sessions in which the individual continuously interacts with the cognitive assessment and training system of claim 7 or claim 8 for at least a predetermined period of time, the cognitive assessment and training system being configured to e the process of any one of claims 1 to 5, wherein; the step of processing the interaction data is med at least before and after the ive training sessions to assess improvements in one or more of the attention-related abilities of the individual.