Update

TRENDS in Cognitive Sciences

Vol.8 No.10 October 2004

437

emphasizes relevant details, even at the expense of a faithful representation of the sensory input.
References
1 Rensink, R.A. et al. (1997) To see or not to see: the need for attention to perceive changes in scenes. Psychol. Sci. 8, 368373 2 Pashler, H.E. (1997) The Psychology of Attention, MIT Press 3 McAdams, C.J. and Maunsell, J.H.R. (1999) Effects of attention on orientation-tuning functions of single neurons in Macaque cortical area V4. J. Neurosci. 19, 431441 4 Treue, S. and Martinez-Trujillo, J.C. (1999) Feature-based attention inuences motion processing gain in macaque visual cortex. Nature 399, 575579 5 Lee, D.K. et al. (1999) Attention activates winner-take-all competition among visual lters. Nat. Neurosci. 2, 375381 6 Lu, Z.L. and Dosher, B.A. (1998) External noise distinguishes attention mechanisms. Vision Res. 38, 11831198 7 Baldassi, S. and Burr, D.C. (2000) Feature-based integration of orientation signals in visual search. Vision Res. 40, 12931300 8 Yeshurun, Y. and Carrasco, M. (1998) Attention improves or impairs visual performance by enhancing spatial resolution. Nature 396, 7275 9 Carrasco, M. et al. (2002) Covert attention increases spatial resolution with or without masks: support for signal enhancement. J. Vis. 2, 467479 10 Carrasco, M. and McElree, B. (2001) Covert attention accelerates the rate of visual information processing. Proc. Natl. Acad. Sci. U. S. A. 98, 53635367 11 Fries, P. et al. (2001) Modulation of oscillatory neuronal synchronization by selective attention. Science 291, 15601563 12 Carrasco, M. et al. (2004) Attention alters appearance. Nat. Neurosci. 7, 308313

13 Tsal, Y. et al. (1994) Attention reduces perceived brightness contrast. Q. J. Exp. Psychol. 47A, 865893 14 Prinzmetal, W. et al. (1997) The phenomenology of attention: II. Contrast and brightness. Conscious. Cogn. 6, 372412 15 Blaser, E. et al. (1999) Measuring the amplication of attention. Proc. Natl. Acad. Sci. U. S. A. 96, 1168111686 16 Sclar, G. and Freeman, R.D. (1982) Orientation selectivity in the cats striate cortex is invariant with stimulus contrast. Exp. Brain Res. 46, 457461 17 Albrecht, D.G. and Hamilton, D.B. (1982) Striate cortex of monkey and cat: contrast response function. J. Neurophysiol. 48, 217237 18 Di Russo, F. et al. (2001) Automatic gain control contrast mechanisms are modulated by attention in humans: evidence from visual evoked potentials. Vision Res 41, 24352447 19 Reynolds, J.H. et al. (2000) Attention increases sensitivity of V4 neurons. Neuron 26, 703714 20 Martinez-Trujillo, J.C. and Treue, S. (2002) Attentional modulation strength in cortical area MT depends on stimulus contrast. Neuron 35, 365370 21 Treue, S. (2001) Neural correlates of attention in primate visual cortex. Trends Neurosci. 24, 295300 22 OConnor, D.H. et al. (2002) Attention modulates responses in the human lateral geniculate nucleus. Nat. Neurosci. 5, 12031209 23 Treue, S. (2003) Visual attention: the where, what, how and why of saliency. Curr. Opin. Neurobiol. 13, 428432 24 Parkhurst, D. et al. (2002) Modeling the role of salience in the allocation of overt visual attention. Vision Res. 42, 107123

1364-6613/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tics.2004.08.001

Algebra and the adolescent brain

New fMRI evidence suggests that adolescents could be at an advantage for learning algebra compared with adults. Qin and colleagues present ndings indicating that after several days of practice adolescents rely on prefrontal regions to support the retrieval of algebraic rules to solve equations, as do adults. Unlike adults, however, after practice adolescents decrease their reliance on parietal regions, which assist in the transformation of the equations, suggesting an enhanced ability for learning algebra. These ndings are discussed with regard to adolescent brain maturation. Have you ever been stumped when helping your teenager with an algebra equation and then realize (with horror) that they actually understand it better than you do? Doesnt it seem odd that although adolescents seem limited in their ability to perform some mental tasks, such as assessing risky behavior, they can be surprisingly adept at others, including complex reasoning such as that
Corresponding author: Beatriz Luna (lunab@upmc.edu). Available online 26 August 2004
www.sciencedirect.com

required in algebra? Using fMRI, Qin and colleagues [1] present intriguing evidence indicating that there may be a brain basis for adolescents ability to implement the high-level logical reasoning required to perform algebraic equations. Cognitive and brain development in adolescence Adolescence is a period when the basic cognitive building blocks that have taken root during childhood are beginning to be rened. As such, the adolescent brain might have unique plasticity for learning. Although salient changes in mental abilities and brain maturation occur in infancy and childhood, there are signicant improvements that continue through adolescence which are largely underappreciated (much like the parent of an adolescent). During adolescence scholarly demands increase dramatically as abstract thought and rule formation become essential to the ability to perform the math and reading required by school curriculums. Executive function, the abilities that include working memory and response inhibition, which allow us to have goal-directed

438

Update

TRENDS in Cognitive Sciences

Vol.8 No.10 October 2004

voluntary behavior, start to mature in adolescence [2,3]. In addition, there are great strides in abstract thought and reasoning. Supporting these enhancements in cognition are changes in brain morphology. Although the gross morphology of the brain is in place by mid-childhood [4], there is continued maturation through adolescence, including synaptic pruning and myelination, which improves the efciency of neuronal computation thought to support the transition to mature cognition [3]. Brain function underlying the solving of algebra equations: adolescents versus adults Qin and colleagues [1] performed an fMRI study to see if adolescents recruit a similar circuitry to adults when learning the complex problem-solving underlying algebraic operations. Their study identied brain regions that showed changes in activation from initial attempts at solving algebra equations to attempts to solve equations following 4 days of practice in 10 adolescent subjects. These results were contrasted with an earlier study in which adult subjects performed algebraic transformations using non-arithmetic symbols (articial algebra) [5]. They found that, similar to adults, adolescents demonstrated increased activity in prefrontal, parietal and motor regions during problem-solving and that prefrontal regions showed decreased activation with practice. Unlike adults, however, adolescents demonstrated decreased activation also in parietal regions with practice. Modeling brain function Although developmental differences in brain function are often evident, the role of specic regions is not always clear. The special contribution of this study was that it addressed the signicance of developmental differences in a particular brain region by using a computer modeling platform called ACT-R (adaptive control of thoughtrational). Earlier studies performed by this group on adults using ACT-R delineated the processes (modules) underlying the problem-solving involved in algebraic equations and the brain regions supporting each of these modules [5,6]. Three modules were identied to be at the core of this process: the imaginal module, which maintains and performs transformations on the equations and is subserved by parietal cortex; the retrieval module, which accesses stored information relevant to algebraic equations and is subserved by prefrontal cortex; and the manual module, which performs the button-press responses required in the experiments and is subserved by motor cortex. Is there a critical period for learning algebra? In the framework of the proposed ACT-R model, the fMRI results indicate that adolescents decrease their reliance on the imaginal/parietal module after they have practised (learned) algebraic equations, whereas adults are still dependent on this module even after practice. These results are very intriguing because they seem to suggest that, as adults, we might be limited in our ability to learn the mental operations underlying this level of problemsolving. Could it be that our teenagers are actually smarter than we are? It is premature at this point to say that
www.sciencedirect.com

adolescence is the critical period for learning higher-order problem-solving, especially because actual performance for the two age groups was equivalent, indicating no advantage to using one circuitry over another. Differences in brain function could also reect brain maturational changes, such as the loss of synapses [7] and myelination [8], known to occur during adolescence. The loss of synapses is believed to underlie reductions in cortical gray matter and is probably a mechanism that allows plastic reorganization of cortex to adapt to environmental demands. Both prefrontal and parietal regions are still maturing in adolescence, as shown by such decreases in gray matter [9] (Figure 1). An incompletely pruned brain could allow plasticity, but would also be limited in the efciency of neuronal computations and integration of brain circuits. The angular (BA39) and supramarginal (BA40) gyri, which were the parietal regions found to show developmental differences in the Qin et al. study [1], are the last parietal regions to mature in adolescence [9]. The results of this study could be interpreted as indicating that the plastic or adaptive aspect of learning high-level problem-solving relies on plasticity in ventrolateral parietal regions. Alternatively, adolescents might be relying more on prefrontal and less so on parietal cortex during practised algebraic calculations to compensate for immature brain structure and limited expertise. Immature parietal circuitry might have a limited role in problem solving, and prefrontal cortex, which has increased participation with increasing task difculty, could be recruited to compensate for limitations in parietal cortex. Thus, reduced activation in parietal cortex following practice could reect an immaturity of this circuitry to support learned behavior. Increased activation in prefrontal cortex could reect perceived task difculty as well as immature circuitry, namely excess synapses. Studies of response inhibition, which does not mature until adolescence [2,10], have indicated that adolescents, who have difculties in suppressing an impending motor response, show increased activation of prefrontal cortex [11]. This provides further support that an immature system might rely on this region to compensate for difculty in accessing the appropriate parietal circuitry. Interestingly, the adolescents in Qin et al.s study also demonstrated increased activation after practice in the putamen. The putamen is a structure of the basal ganglia that receive inputs from motor and somatosensory cortex and has projections to the supplementary motor area, a region that supports response planning. Increased activation in the putamen supports the notion that adolescents might need increased effort to perform complex calculations at mature levels, relying on brain regions that are not necessary in adult performance. The basal ganglia have also been found to show protracted maturation [12]. This region was not revealed in the computer simulation that was based on adult problem-solving and might be exclusive to learning in the immature adolescent system. Future directions This study lays the groundwork for further investigation. First, it would be important to know whether activation

Update

TRENDS in Cognitive Sciences

Vol.8 No.10 October 2004

439

5y

rs

Ag

>0.5 0.4
20 yrs

0.3 0.2 0.1 0.0 Grey matter volume

Figure 1. Right lateral and top views of the dynamic sequence of gray matter maturation over the cortical surface. The side bar shows a color representation in units of gray matter volume. The supramarginal gyrus is the last area of parietal cortex to develop, between the third brain from the left (approximately 13 years of age) and the fourth (approximately 16 years of age). Reprinted with permission from [9]. Copyright (2004) National Academy of Sciences, USA.

differs between adults and adolescents performing the same articial algebra in the same study. Comparing results from adults and adolescents from different studies undermines the power of comparison between age groups. A study with both adults and adolescents would allow direct comparison of brain regions between age groups so that degree of activation in different regions could be compared. Although adults do not show a reduction of parietal activation with practice, they might be tapping into parietal regions in a similar way to the adolescents recruitment after practice. Alternatively, if adults were allowed to practise for a longer time perhaps they would demonstrate the decrease in reliance on parietal regions that adolescents show. This approach would help to establish whether the differences in brain function are related to using real versus imagined algebra or are in fact related to age differences. Gender differences would also be of interest given the discrepancy in math scores between males and females. fMRI could potentially indicate brain differences that might limit females in their ability to learn such problemsolving at the same age as males. Finally, only 10 adolescents were tested in the Qin et al. study. A larger number of subjects would assist in accounting for intersubject variability that might itself be related to age given that adolescents reach maturity at different ages. What is evident though is that teenagers and adults are processing practised information differently, perhaps helping to explain why we adults feel rusty when performing these equations.

References
1 Qin, Y. et al. (2004) The change of the brain activation patterns as children learn algebra equation solving. Proc. Natl. Acad. Sci. U. S. A. 101, 56865691 2 Luna, B. et al. Maturation of cognitive processes from late childhood to adulthood. Child Dev. (in press) 3 Luna, B. and Sweeney, J.A. The emergence of collaborative brain function: fMRI studies of the development of response inhibition. Ann. N. Y. Acad. Sci. (in press) 4 Caviness, V.S. et al. (1996) The developing human brain: a morphometric prole. In Developmental Neuroimaging: Mapping the Development of Brain and Behavior (Thatcher, R.W. et al., eds), pp. 314, Academic Press 5 Anderson, J.R. et al. (2003) An information-processing model of the BOLD response in symbol manipulation tasks. Psychon. Bull. Rev. 10, 241261 6 Qin, Y. et al. (2003) Predicting the practice effects on the blood oxygenation level-dependent (BOLD) function of fMRI in a symbolic manipulation task. Proc. Natl. Acad. Sci. U. S. A. 100, 49514956 7 Huttenlocher, P.R. (1990) Morphometric study of human cerebral cortex development. Neuropsychologia 28, 517527 8 Yakovlev, P.I. and Lecours, A.R. (1967) The myelogenetic cycles of regional maturation of the brain. In Regional Development of the Brain in Early Life (Minkowski, A. ed.), pp. 370, Blackwell Scientic 9 Gogtay, N. et al. (2004) Dynamic mapping of human cortical development during childhood through early adulthood. Proc. Natl. Acad. Sci. U. S. A. 101, 81748179 10 Munoz, D.P. et al. (1998) Age-related performance of human subjects on saccadic eye movement tasks. Exp. Brain Res. 121, 391400 11 Luna, B. et al. (2001) Maturation of widely distributed brain function subserves cognitive development. Neuroimage 13, 786793 12 Sowell, E.R. and Jernigan, T.L. (1998) Further MRI evidence of late brain maturation: limbic volume increases and changing asymmetries during childhood and adolescence. Dev. Neuropsychol. 14, 599617
1364-6613/$ - see front matter Q 2004 Elsevier Ltd. All rights reserved. doi:10.1016/j.tics.2004.08.004

www.sciencedirect.com