variability, attention, and strategy use.

Variability - CEP 903 11.9.07

• Variability exists between gifted and mainstream children, as illustrated by Johnson et. al. (2003). In their research, younger and older gifted children and mainstream children were tested on M capacity tasks and tasks measuring speed of processing, inhibition, and interference. Results obtained show that gifted children possess superior mental attention capacity, perform higher on speeded tasks, and are better at controlling attention (resisting interference). Consequently, greater mental capacity results in an increase in cognitive resources.

Education implications would tell us that because gifted children focus better, they are less likely to be distracted in class whereas mainstream children would have more difficulty staying on-task in the presence of distraction. Results obtained in this study could suggest that gifted children may have learned to form automatic associations between two concepts (i.e. trail making task – they associate 1 with A, 2 with B, etc) in a way that increases the likelihood that activation of one idea will prime activation of the other idea (1 primes A, 2 primes B, and so on). If this is the case, then they may not truly be better at effortful inhibition, rather, they are better at making associations that allow inhibitions to become automatic.

Gifted children also did not show superior inhibition, suggesting that they were not better at suppressing task-irrelevant information. This finding is illustrated by negative priming and shows that when a distracter stimulus in one trial becomes the target stimulus on the next trial, gifted children had slower response times because previous inhibition dampened the activation level of the stimulus. ADHD children, on the other hand, probably shift attention well, but do not focus well; therefore, they would be less influenced by negative priming.

• Another study on attention illustrates a second instance of variability in developing children. Jones et. al. (2003) conducted a study in which 3 and 4 year olds were tested on their ability to follow or inhibit instructions given by two different sources in a Simon Says task. Results show that younger children had lower accuracy whereas older children made few inhibitory errors. However, when the older children did make an error, their reaction time on the next trial was long. This could be perceived as an early sign of metacognition, as defined by Miller et. al. (1986). Older children who made few inhibitory errors were trying to figure out what they did wrong while younger children were unable to detect their own errors. As the results would suggest, there is a negative correlation between attentional shifting and ability to inhibit responses.

Educational implications from this article would suggest that preschoolers who shift attention easily have difficulty maintaining focus, which could be an indication of ADHD or some type of attentional problem. ADHD children would lose focus over fewer trials and would experience less trial to trial interference because they would be less aware if they made a mistake. These results are related to Kruschke’s work (2003) in that they suggest that attention to a target requires inhibition of a distracter.

Because accuracy to error-related performance is determined by brain structures in the frontal cortex, development of attention should depend on proper development of frontal brain structures (Nelson). For example, a negative prenatal environment can lead to stunted brain development, which would ultimately be detrimental to one’s ability to reduce error in attention inhibition tasks when compared to normally developing children.

• Strategy development as studied by Miller et. al. (1986) and Welch-Riss et. al. (2000) reveal variability in 4, 6, 8, and 10 year old children by testing their ability to use strategies effectively. Younger children were less likely to use a strategy or relied on one strategy, whether it worked or not. Older children, on the other hand, were not only more likely to use a strategy, but used a strategy tailored to the task.

Results would argue that increase in age leads to increase in automaticity of strategies for several reasons. First of all, neural connections become stronger with experience (Nelson). Secondly, as illustrated by Johnson et. al. (2003) and Jones et. al. (2003), younger children are less able to utilize multiple cognitive sources (i.e. memory and strategy). Finally, in agreement with Johnson and Jones and as stated by the limited capacity model (Miller et. al., 1986), capacity to access and use the right strategy and monitoring its use leaves little capacity for memory (TOM). Although age contributes to strategy use, so does type of strategy used. Experience is key for becoming automatic at strategies and becoming metacognitive (illustrated by the deception task).

Educational implications include realizing that young children do not have enough cognitive space to attend to multiple things at once and are less likely to use strategies because it requires memory, a resource that is already exhausted because mental capacity is already being applied somewhere else. As Kruschke would infer (2003), what we remember is limited by attention: if we pay attention to A, we have fewer resources available to attend to B.

• Finally, literacy achievement as described by Reynolds et. al. (1990) show variability in children as they develop. They found differences between less successful and more successful 10^th grade readers. More successful readers were likely to use selection attention strategies (SAS) to attend to important information and were better at conceptual tasks. Less successful readers, on the other hand, were less likely to attend to important information because their attention was directed toward perceptual qualities.

Results would suggest that attention does not necessarily equal learning and perceiving does not mean processing if attention and perception are focused on irrelevant properties. Increased metacognitive awareness allowed the more successful readers to utilize SAS to recall the important information.

These findings have several educational implications for literacy achievement. To begin with, they suggest that using appropriate strategies leads to better learning (Siegler, 2007). They also suggest that early reading is important because once language pathways lose stimulation, those pathways are more difficult to redirect and strengthen (Nelson). In order for readers to comprehend, decoding and vocabulary have to become automatic by 1^st grade (Juel, 1988).

Domain specific vs. domain general

Pinker proposes that language development is domain specific (1994). In other words, he believes that our abilities are prewired and controlled by separate mechanisms, processes, and pathways. His research is supported by instances in which individuals with language impairments can possess intact intellectual abilities while individuals with impaired intellectual abilities can possess intact language. He states that grammar structure is universal and that brain structure for language is the same in everyone. His argument may be supported by Nelson who states that the brain creates and strengthen new, different pathways in the brain to activate language competency in someone with a speech and language impairment.

Other research suggesting that learning is domain specific is illustrated by Reynolds et. al. (1990). Their research shows that children can have intact decoding abilities (perception) but impaired comprehension abilities (conception). They suggest that effective versus ineffective attention strategies determine reading success or failure and conclude that reading ability is controlled by separate mechanisms.

Smith would argue that language development is domain general and that learning is experience dependent (1999). General purpose mechanisms and associative learning are means by which we learn. Smith stated that words initially have no meaning until they are associated with something. According to the shape bias, children attended to shape in naming objects because as children learn words, the act of naming becomes a contextual cue that automatically recruits attention to shape.

Siegler would also agree that learning is domain general. According to his research, we possess general learning mechanisms that are refined by variability outcomes. Periods of stability (low variability) alternate with periods of transition (high variability) that help us determine which strategies are effective and ineffective. Learning is most likely when previous strategies weaken (due to failure, negative feedback) and new, more efficient strategies strengthen (i.e. infant reaching as described by Smith and Thelan, 2003).

I find the domain general learning mechanism arguments more compelling because they suggest that learning is experience dependent. If learning is prewired and innate, then education and rehabilitation would be useless. Or genes are expressed differently through different experiences and increases in experience lead to confirmation of certain probabilities (Saffran, 2003). Learning must be domain general because otherwise, associative learning would not occur. According to Nelson, learning is experience dependent; therefore, environmental influences must play a role. If a child with a brain impairment was viewed as unable to change, then what would be the point of teachers, speech pathologists, occupational therapists, physical therapists and so on? Likewise, how would we explain individual variability? If our brains were composed of specific mechanisms that direct learning, then we would be more alike than different.

The only way to reconcile these different perspectives is to realize that nature and nurture both contribute to learning. Our learning is experience dependent; however, our range of reaction and early experiences put constraints on later change and development of our abilities. Our brains do contain structure, but we all rely on external support for brain functioning. Infant brains, for example, have greater plasticity and are therefore more susceptible to learning. We cannot assume that all of our learning is due to domain specificity or domain generality because both are crucial to our development.

cognitive learning mechanisms

Teaching reading - CEP 903 10.5.07

• Statistical learning, as introduced by Saffran (2003), suggests that language is shaped by human learning mechanisms rather than innate brain structures. We learn to predict statistical probability by becoming familiar with sequences in language. For example, presenting the word ‘the’ or ‘a’ in text serves as a cue, helping us predict that a noun will follow it. Therefore, an increase in language experience increases our ability to confirm certain language probabilities. Teachers can utilize Saffran’s work when teaching reading to students by exposing them to print at an early age, focusing on language concepts such as phonetic features, word boundaries, and syntax.

• Embodiment refers to a concept claiming that we learn in our bodies and interact with our environment to obtain knowledge. Our behavior is a result of encoding perceptions. In a study conducted by Noice and Noice (2001), for example, movement facilitated recall even without intent because environmental cues provided scaffolding. Their finding suggests that active experiencing helps create meaning and enhances memory because real actions have properties that language descriptions do not. Teaching reading that utilizes embodiment could include role playing parts of a story or using concrete objects and hands-on experiments to help students visualize abstract concepts (Noice and Noice, 2001; Gentner, 2002).

• Test-enhanced learning suggests that testing students on material is more effective for retaining information in long term memory than repeated studying of the material (Roediger, 2006). Studying (or rereading) is different than retrieving information from our brains (testing) because the latter allows us to practice the skill actually required on future tests. Knowing this, an effective approach for teaching would be to expose students to reading material and administer periodic testing over the material. Students who are taught by this method would have the opportunity to demonstrate what they have learned on multiple occasions, each time strengthening their knowledge of the material.

• Maternal elaboration is found to effect child elaboration and memory over time (Reese, 1993). Utilizing elaborative techniques provides scaffolding for the learner in which he/she is able to re-experience learning the information each time a new elaboration is introduced. Reese’s findings provide implications for teaching in that they suggest effective ways for teachers and students to interact. Teaching by elaboration rather than repetition would provide bidirectional scaffolding that ultimately increases teacher-student interaction, facilitating communication and recall of information. In other words, asking a variety of questions over reading material allows the learner to make more connections in the brain because more resources are being activated whereas repeating the same question often does not provide adequate scaffolding for the learner.

• Glenberg’s memory model describes memory as being embodied by combining different sets of actions together (1997). In order to remember something, we either use clamping to ignore memories and attend to the environment or use suppression to decrease our current perception and enhance memories. Limitations on learning according to Glenberg are based on possible next steps that we know to be true because of our experiences. For example, in repetition priming of language, our previous exposure to reading material facilitates our current ability to process it. Teaching implications that utilize repetition priming are related to our ability to statistically predict language (Saffran, 2003) and mesh concepts (Glenberg, 1997). Teaching students to remember what they read can be modeled after repetition priming by creating multiple opportunities for students to associate concepts by meshing related words. For example, presenting the word ‘volcano’ and allowing students to choose from a list of words that are related to the word primes ‘magma’ or ‘lava’ but not ‘dog’ or ‘cat.’ ‘Magma’ and ‘lava’ are likely stored in the same category in the brain whereas ‘dog’ and ‘cat’ are not; therefore, memory retrieval is strengthened.

Empirical evidence

Brain research provides compelling evidence for learning and development in that it gives specific examples of how bidirectional change affects brain development (Nelson, 1997). As a basic explanation of Nelson’s findings, we know that learning and thinking equal brain change. Bidirectional change can be described as experiences that lead to brain change and a changed brain that leads to changed experiences. For example, an infant brain possesses great capacity for change because neurons in the brain are initially uncommitted. Once connected, the neurons become stronger and neuroplasticity decreases. As a result, each experience sets limits for later change.

Infancy can be considered a foundation for future learning in which boundaries for brain development are set by a range of reaction. An infant’s range of reaction puts constraints on intellectual ability and positive or negative results are manifested by environmental influences. Range of reaction could be determined by a number of factors. Stressful pregnancy and malnutrition are biological influences that contribute to decreased brain development. Poor parenting and lack of exposure to educational materials are environmental influences that contribute to negative bidirectional change. In each case, fewer opportunities are available for high intellectual ability.

Brain research has clear implications for education in that it gives teachers hope that the brain has plasticity. At the same time, it supports the notion that parents have a crucial role in shaping healthy early learning experiences. Furthermore, it allows teachers to understand why children have different intellectual abilities and behave differently. Brain research confirms that early intervention is more effective than later intervention because it provides more opportunity for positive brain change and experience.

Unlike the convincing evidence found in Nelson’s research, Reese’s evidence suggesting that maternal elaboration is an effective technique for facilitating childrens’ memory is less compelling (1993). According to Reese, highly elaborative mothers facilitate child elaboration and memory of information. The evidence supporting this claim argues that elaboration creates scaffolding for recalling past events and that a child is able to contribute more information about these events if the mother is elaborative.

The first problem with these findings is that they are ungeneralizable. The study results show that high elaborative style is associated with girls more than boys. If boys are exposed to less elaboration, they will inevitably become less elaborative as an adult. In other words, socializing girls to elaborate more begins as a small difference that becomes magnified over time. As a result, dads are likely to be much less elaborative. This brings me to my next criticism. Father elaboration style was not addressed perhaps because it would fail to support that the concept of high elaboration style is generalizable. This study also fails to take environmental factors into consideration. Ability to use elaboration clues may vary according to a child’s ability to block out distracting information in the environment. If memory is enhanced by scaffolding, it seems that it should be worsened by irrelevant cues in the environment. Finally, misleading elaborations may foster the recall of inaccurate information; therefore, high elaboration style from a misinformed source may hinder rather than help recall accurate information.