cognitive fun!

Okay, does the dual n back task really work, or is this just a publication bias

2010 study shows no improvement on Raven's w/ training

"These findings can be addressed from the more parsimonious view that CWM training affects reading comprehension in the same way that it benefits both WM (verbal and spatial) and Stroop performance: by enhancing a domain-general attentional mechanism not specifically tied to verbal or spatial storage per se, but to the coordination of information maintenance in the face of additional processing demands. This interpretation is consistent with the broader claim that CWM tasks are highly predictive of cognitive function specifically because they limit rehearsal and domain-specific strategy use and, hence, emphasize the role of domain-general attentional mechanisms (Cowan et al., 2005). Since verbal rehearsal strategies are so highly developed, the suppression of rehearsal and the isolation of attention control mechanisms may be more complete in spatial CWM span tasks (Conway et al., 2005). Accordingly, the inclusion of a spatial CWM component in training may have been especially important in producing the observed generalization effect. Indeed, our correlational findings suggest that reading comprehension improvements were particularly dependent on increased spatial WM performance."

"Admittedly, one difficulty for the claim that transfer arises from the impact of WM training on a domaingeneral attentional process is that both improved and unimproved measures included in our assessment battery have been linked, on the basis of latent variable studies, to this shared WM mechanism (Kane et al., 2004). The selectivity of transfer invites the question of why training generalized to some of the cognitive skills that we measured, but not to others. Specifically, the training group did not show differential improvement on measures of reasoning or fluid intelligence, which also putatively demand controlled attention. One way to address the apparent discrepancy is to assume that tasks for which we observed improvements simply place a higher demand on controlled attention than do the unaffected tasks. Another possibility is that our study lacked the statistical power to reveal modest training effects in seemingly unaffected measures. Indeed, there was a trend toward significance in the effect of training on the ETS reasoning composite that would likely have reached statistical significance in a larger sample."

http://findarticles.com/p/articles/mi_7721/is_201004/ai_n53507996/pg_7/?tag=content;col1

I'd say the odds against Jaeggi etc. claim of enhanced iq are 2:1. However unlike the BBC study, Chein and Morrison conclude that there is enhancement of reading comprehension-meaning that dual n back training is probably worthwhile.