Processing speed tasks: There were three processing speed tasks; verbal baseline processing, visual choice, and digit cancellation. In each episode of each task, there were 20 test trials, plus 5 practice trials, with no trials duplicated between or within sessions. For each task, children completed two episodes, the first in Session 1, and the second in Session 3. In the verbal baseline processing task children were required to decide whether a word presented auditorily to them was an animal name by pressing a button on the laptop keyboard. Words were assigned to two sets (animals or non-animal concrete nouns). Words were matched between sets for number of syllables and age of acquisition. The task moved on when 500ms had elapsed after a decision key had been registered. The visual choice task required children to rapidly press one of two keys, each of which had been assigned to a picture of one of two different frogs. The frogs were visually very different and their pictures were chosen to be difficult to verbalise with any descriptor other than ÔfrogÕ. When the a frog image appeared in the centre of the screen, children were required to press the corresponding key. The task moved on to the next trial when 300ms had elapsed after the decision had been made. In the third task, digit cancellation, the child was presented with a screen containing 20 buttons, arranged in grid pattern, each showing a number between 1 and 9. The task was to find the number 5s on screen as quickly as possible, of which there were five. Children searched for the Ô5Õs using an external computer mouse, as quickly as possible, and clicked on the buttons they chose. Finding a 5 rewarded the child with a smiley face icon, which appeared in place of the button. If an incorrect button was clicked, the button remained on screen and no icon was shown. When all of the 5s had been found on one screen, another screen was presented until the child had completed 4 screens (and had therefore made a total of 20 correct clicks). The time taken to click each correct button was recorded. Digit span task: Children were presented with increasing lists of digits (two to eight) with five trials at each list length. If children successfully remembered any one list in correct serial order from the five trials at a given list length they moved on to the next list length. If they failed to recall all lists within a given list length, testing was terminated at that point. Children were presented with a digital audio recording of digits in a male voice and a 3 cm high Arabic numeral, in black, for 1,000 ms in the centre of the computer screen. A blank screen appeared briefly between each word. At the end of each trial, a cartoon character appeared alongside a question mark, prompting the participant to recall the words in the order they had heard them. Both span (the longest length of list that could be recalled in correct serial order on any trial), and partial credit score were recorded. Under the partial credit score method, proportional credit is given for each item recalled at the correct serial position in any list. For example, in a list of three items, each item has a potential proportional score of 0.333. Summed proportional scores are then totalled across all trials within each span task. Complex span task: This task followed the span procedure used in the digit span task (with the same number of trials at each of the same list lengths, and the same continuation and stopping criteria). Children with digit spans below 4 items started testing at 2 item trials, while children with higher digit spans began testing at 3 item trials to save time. When started on a 3 item trial, if no trials were correctly recalled at that level, testing then moved backwards to 2 items. Digits were presented in a male voice as digital audio recordings, simultaneously with the appearance of the numeral being shown in black in the centre of the computer screen measuring approximately 2 cm high for 1,000 ms. Between each digit presentation one processing episode occurred, which the children had unlimited time to complete, which was one of the verbal baseline processing decisions (is the word an animal or not?). When a decision was made, there was a pause of 750ms before the task moved on to the next digit in the memory sequence. When the full sequence of digits had been presented, children were asked to recall the sequence in correct serial order. All processing episodes used in the complex span task were unique to the task and had not been used in any other task (70 unique words were employed in the processing phases of this task. Immediate free recall task: Participants were presented with word lists spoken in a male voice ranging from 5 - 8 items in length, with four trials at each list length, giving a total of 20 trials. List lengths were pseudo-randomly organized in five testing blocks, so that list length was unknown to the child before presentation of any given list. Children were presented with a cartoon character and told that he had words for them to remember. The word lists were then presented with the character in the corner of the screen and a speech bubble coming from his mouth in the centre of the screen. An audio recording of each word simultaneous with a 3 cm high color illustration of the word was presented in the centre of the speech bubble for 1,000 ms, followed briefly by a blank screen and the next word. At the end of each trial, a question mark appeared above the cartoon character's head, and children were asked to recall as many words as they could, in any order. Interleaved lists task: Children were introduced to two cartoon characters, who were identified by separate illustrations and two distinct male voices. They were explicitly told to pay attention to the words spoken by Character A (focal or ÔattendedÕ stimuli) and to try to remember his words in any order, and to try to ignore words spoken by Character B (non-focal items). Two total list lengths of 4 and 6 items were presented. Focal items were always presented first in sequence, and focal and non-focal items were interleaved with one another, for example, the presentation order for a 4 item list was focal Ð non-focal Ð focal Ð non-focal, (6 item lists had 3 focal items and 3 non-focal items interleaved). Items in each sequence were presented pictorially with a color cartoon image of the word, accompanied by an audio recording of one of two different male voices (one for Character A and one for Character B). Character A always appeared on in the left bottom corner of the screen and Character B on in the right bottom corner. Images of the stimuli were presented in the centre of the screen in a speech bubble originating from the relevant character's mouth, and were displayed for 1,000 ms, with a 250ms pause between each word. Children were only asked to recall focal (attended) items in order to maximize the chances of participants maintaining just the focal items in primary memory. There were five trials at each list length, giving a total of ten trials, and these were pseudo-randomly organized. After presentation of each trial, Character AÕs speech bubble reappeared in the centre of the screen, highlighted in red alongside Character A, which signalled that the child should try to recall the focal words only. Forgetting task: The forgetting task was set at a level that was calibrated to an individualÕs span level (determined by the digit span task), meaning that each child was presented with lists of digits at the span level which they could recall correctly in sequence. Children were presented with digit sequences followed by as many processing decisions as they could complete within a set time period. Time periods were set at 2, 4.5, 7.5, 11, and 15 seconds, and a version of the verbal baseline processing task with 70 new and unique words (see above) was used as the filler task. If the specified total time had elapsed prior to a final decision being made, the processing window remained open until the children had completed their final decision. Children were then asked to recall the initially presented digit sequence in serial order, and their output order was recorded. Reaction times to the processing elements of the task were also recorded by the computer. There were 3 trials of each time period, giving a total of 15 trials, and 2 further practice trials. Trials of each time period were presented in a pseudo-random order to ensure that no two consecutive time periods were the same, and that time periods were not presented in an increasing or decreasing order. Partial credit scoring was used to calculate forgetting task performance at each time point. However, in this task, a forgetting rate value was calculated in order to determine the extent to which individualsÕ retention had dropped relative to their memory for the same length lists of digits with no delay. This was calculated by summing partial credit scores for just the trials of the digit span task that were presented at span level. The free recall partial credit scores from forgetting trials at each time point (2, 4.5, 7.5, 10, and 15 seconds) were then proportionalised against this starting point. The sum of these values gives a forgetting score for this task. Memory period task: Children were presented with mathematical problems that they had to solve in order to find the correct memory item (e.g. 2+1 produces a memory item of 3). There were 3 trials at each level, ranging from level 2 to 5. Each level had 4 memory items to remember but the number of operations presented varied at each level, increasing the amount of time to potentially forget information. At level 2, 2 operations had to be completed (e.g. 4+1+0) while at level 5, 5 operations had to be completed to access the answer (e.g., 4+1+0+3+1-2). Children had as much time as they needed to complete any sum presented. The experimenter recorded whether responses to sums were correct or incorrect, and processing time was recorded for each sum by the experimenter clicking on a computer keyboard button at the onset of responding. After a response to the sum had been recorded by the experimenter, the programme moved on to the next sum after 500ms. After 4 sums, or Ôoperation periodsÕ, children were requested to recall the memory items in correct serial order. Administration followed a span procedure, where the task was terminated if a child could no longer recall all trials at that level in the correct serial order. Children who reached higher levels were therefore more able to retain 4 memory items under increasing periods of distraction. The dependent variables from this task were the initiation time to begin recall of memory items, and partial credit score for recall Timed articulation tasks: Children were asked to repeat a single word, or a word pair, five times, as quickly as they could. There were six single words (cat, pig, bin, rug, bug, rope) and six word pairs (cat-rug, pig-bin, pig-rope, bug-rope, rug-bin, cat-bug) and responses were recorded on a digital voice recorder. Children were introduced to the task in 2 practice trials, and were allowed to repeat the practice trials until they felt able to complete the task. Single words were always presented before word pairs. Average times taken to utter 5 repetitions of each single word and 5 repetitions of each word pair was were recorded. Responses more than 2.5 standard deviations above or below the mean were replaced with the cutoff value for that set of five repetitions. Delayed free recall task: In this task, five trials of 15-word lists were presented. In each trial, an audio recording of each word simultaneous with a 3 cm high color illustration of the word was presented in the centre of the computer screen for 1,000 ms, followed briefly by a blank screen and the next word. Each list featured an exemplar word from each of 15 predetermined categories. After a word list had been presented, the participant completed a distracter task. In the distracter task, the participant was presented with 20 buttons (1cm diameter) arranged in a grid pattern on a new screen, which featured the numbers 1-20. The child was required to click all of the buttons on the screen in their correct number order as quickly as possible (20 buttons displayed), until no more remained on screen. When the child clicked a button, a smiley face icon appeared on screen in place of each button. When they had completed this distracter task, the child was then asked to what they could remember from the previously presented word list. There were three phases to this recall period. In the first, they were asked to spontaneously recall as much as they could from the previously presented list. In the second, they were asked to recall the words which was had been presented from four probe categories, not previously recalled in the first phase (e.g., item of clothing, family member, etc.). In the third phase, eight additional probe words were presented to the child, four of which had been selected from the just- presented list that had not been recalled in either of the earlier recall phases, and the child was asked if they had been present or absent in the just presented list. Four foils were interspersed with the probe words which were derived from different semantic categories to the true probe words. There were three dependent variables taken from this task; average number of words across the five trials spontaneously recalled, total recalled with semantic probing (20 possible recalls), and A-prime accuracy in response to the probe words. Final free recall task: At the end of the first session (approximately 20 minutes after the immediate free recall task was completed) children were asked if they could recall anything from the immediate free recall task. The total number of items recalled was recorded. ConnersÕ 3 Teacher Short Form: Class teachers completed the ConnersÕ 3 Teacher Short Form, rating the child's behaviour in the past month. This form consists of 39 statements assessing five sub-components of classroom behaviour: inattention, hyperactivity/impulsivity, learning problems/executive functioning, defiance/aggression, and peer relations. Higher scores reflect a greater prevalence of behaviours. Reading task: A sentence completion task, designed for the current study, was presented which consisted of 30 sentences. The task was a cloze task, in which incomplete sentences were presented which had one word missing. Ten sentences had the missing word at the beginning of the sentence, ten in the middle, and ten at the end of the sentence. For each sentence, children were given five words to choose from as candidates for the missing word, only one of which made both grammatical and semantic sense. Two of the words in each response set were semantic foils. Mathematics task: The Numerical Operations subtest of the Wechsler Individual Achievement Test Second Edition (WIAT-II UK) was used, in which children complete as many numerical operations (counting, pencil and paper computations) as possible. The raw score was used as the dependent variable.