Do brain training software programs and “brain games” work?

uTo deter­mine if some­thing works we first need to define what we mean by “work”. A machine to train abdom­i­nal mus­cles prob­a­bly won’t “work” if what we mea­sure is blood pres­sure. In the same way, a pro­gram train­ing audi­to­ry pro­cess­ing speed may not work if visu­al func­tions are mea­sured (see Chap­ter 1 for a bet­ter under­stand­ing of cog­ni­tive abilities).  This is why to deter­mine whether a brain train­ing soft­ware “works” it is cru­cial to (a) under­stand the claims made by the devel­op­er as to what abil­i­ties are trained, (b) find well con­duct­ed stud­ies show­ing that these abil­i­ties are indeed trained by the pro­gram and © decide whether that train­ing is rel­e­vant to one’s needs and objectives.

Anoth­er impor­tant aspect when eval­u­at­ing whether a brain train­ing pro­gram “works” is to look at the extend to which the train­ing effects trans­fer to untrained tasks. It is well estab­lished that prac­tice usu­al­ly trig­gers improve­ment in the prac­ticed tasks. So the first require­ment for a well work­ing brain train­ing pro­gram is to show that peo­ple who use the pro­gram get bet­ter at the tasks trained. The sec­ond and more impor­tant require­ment is to show that this improve­ment trans­fers to oth­er, untrained, tasks, most­ly tasks per­formed dur­ing every­day life. This would show that the cog­ni­tive abil­i­ties tar­get­ed by the pro­gram were indeed trained. If I use a train­ing pro­gram to train my abil­i­ty to con­cen­trate (atten­tion skills) I will prob­a­bly get bet­ter at the tasks includ­ed in the pro­gram if I prac­tice long enough, but will I see any ben­e­fits when I do oth­er tasks, at work for instance?

Teams of neu­ro­sci­en­tists and psy­chol­o­gists from around the world have part­nered with soft­ware and game devel­op­ers to bring tar­get­ed brain fit­ness prod­ucts to mar­ket with more sol­id clin­i­cal val­i­da­tion. These teams have pub­lished results using the gold stan­dard of ran­dom­ized con­trolled tri­als, sup­port­ed by neu­roimag­ing. As a result, they have been able to claim quan­tifi­able short-term and long-term improve­ments to spe­cif­ic cog­ni­tive skills if used accord­ing to a spe­cif­ic reg­i­men over a spec­i­fied length of time.

Much of the hope and media cov­er­age of the brain fit­ness mar­ket in 2007 can be traced to the pub­li­ca­tion of the results of the five years ACTIVE study con­duct­ed by Willis and her col­leagues (2001, 2006). This study was one of the first ran­dom­ized con­trolled, sci­en­tif­i­cal­ly sound stud­ies ever pub­lished in the area of brain train­ing. Par­tic­i­pants in this ground­break­ing study were 73.6 years old on aver­age. They were exposed to dif­fer­ent forms of men­tal train­ing: rea­son­ing, mem­o­ry and speed train­ing. Strate­gies and prac­tice were pro­vid­ed dur­ing train­ing. The train­ing of pro­cess­ing speed was com­put­er-based. Par­tic­i­pants showed an improve­ment in the skills trained and retained a sig­nif­i­cant per­cent­age of this improve­ment when test­ed five years lat­er. Inter­est­ing­ly, the group who received the train­ing in pro­cess­ing speed showed the most pro­nounced short-term and long-term improvements.

Since the pub­li­ca­tion of the ACTIVE study, a grow­ing num­ber of ran­dom­ized con­trolled stud­ies are show­ing how well direct­ed train­ing soft­ware may pro­duce cog­ni­tive and oth­er improve­ments to dai­ly life.

For instance, in 1994, Dr. Daniel Gopher and his col­leagues used Space Fortress, a com­plex com­put­er game pre­cur­sor to Intel­li­Gym, to train flight cadets. Par­tic­i­pants received 10h of train­ing. Results showed that com­pared to a no-train­ing group, flight cadets who were trained showed a 30% improve­ment in their flight per­for­mance. This sup­ports the idea that the ben­e­fit gained through prac­tic­ing a com­put­er game can trans­fer to a task involv­ing sim­i­lar cog­ni­tive abil­i­ties (flight per­for­mance here).

Cogmed is a com­put­er­ized pro­gram aimed at improv­ing work­ing mem­o­ry (WM). WM is the mem­o­ry sys­tem that allows one to hold infor­ma­tion briefly in mind for the pur­pose of the task at hand. In 2005, Dr. Torkel Kling­berg and his col­leagues con­duct­ed a ran­dom­ized, con­trolled, study to test whether the use of Cogmed could help improve the WM per­for­mance of chil­dren with atten­tion-deficit/hy­per­ac­tiv­i­ty dis­or­der (ADHD). The train­ing peri­od was at least 20 days. Results showed that train­ing WM using Cogmed increased the per­for­mance of the chil­dren in untrained tasks mea­sur­ing WM as well as in tasks mea­sur­ing response inhi­bi­tion and com­plex rea­son­ing. The ben­e­fits were still present when the chil­dren were test­ed again 3 month after the training.

Dr. Arthur Kramer, whose inter­view you can find at the end of Chap­ter 2, just pub­lished the results of a study test­ing the ben­e­fits induced by play­ing a strat­e­gy-based videogame (Rise of Nations Gold Edi­tion). Results of this study showed that trained par­tic­i­pants (age 69 on aver­age) not only got bet­ter at play­ing the game but also showed trans­fer of ben­e­fits to untrained tasks that engaged the same abil­i­ties (work­ing mem­o­ry, task switch­ing, etc.) as the game.

In 2003, Dr. John Gabrieli and his col­leagues used Fast For­word, a com­put­er­ized pro­gram designed to train audi­to­ry pro­cess­ing with 20 chil­dren with dyslex­ia. Func­tion­al Mag­net­ic Res­o­nance Imag­ing (fMRI) was per­formed on the chil­dren dur­ing phono­log­i­cal pro­cess­ing before and after the train­ing. Behav­ioral­ly, the train­ing improved oral lan­guage and read­ing per­for­mance. Phys­i­o­log­i­cal­ly, chil­dren with dyslex­ia showed increased activ­i­ty in mul­ti­ple brain areas cor­re­lat­ed with improve­ment in oral lan­guage abil­i­ty. These results sug­gest that the train­ing improved the tar­get­ed brain func­tions. Although the num­ber of par­tic­i­pants in this study was small, the results were repli­cat­ed in a fur­ther study pub­lished in 2007.

Posit Sci­ence Clas­sic is anoth­er com­put­er­ized pro­gram, designed to train audi­to­ry pro­cess­ing. One con­trolled, ran­dom­ized, study pub­lished in 2006 by Dr. Michael Merzenich and his col­leagues showed that adults age 60 and over trained using Posit Sci­ence pro­gram improved in audi­to­ry tasks. More impor­tant­ly the improve­ment gen­er­al­ized to an untrained mem­o­ry task. The mem­o­ry ben­e­fits were still present 3 months after the train­ing. The same com­put­er­ized pro­gram is ongo­ing fur­ther test­ing in the IMPACT study con­duct­ed by Dr. Eliz­a­beth Zelin­s­ki (see her inter­view at the end of Chap­ter 2). The ini­tial results of this study were pre­sent­ed at the Geron­tol­ogy Soci­ety of Amer­i­ca in 2007. They showed sig­nif­i­cant gains in audi­to­ry pro­cess­ing and audi­to­ry mem­o­ry “equiv­a­lent to ten years of aging for that skill” in over 500 adults with a medi­an age of seventy-five.

In sum, these stud­ies show quan­tifi­able improve­ments to spe­cif­ic cog­ni­tive skills if the test­ed brain train­ing soft­ware is used accord­ing to a spe­cif­ic reg­i­men over a spec­i­fied length of time. In some stud­ies, trans­fer of ben­e­fit to untrained tasks has also been observed.

How­ev­er, it is still too ear­ly to tell whether or not these prod­ucts will result in mea­sur­able long-term health ben­e­fits, such as bet­ter over­all brain health, or low­er inci­dence of Alzheimer’s symp­toms. One of the rea­sons for this, to be fair, is the fact that most of the com­mer­cial­ly avail­able prod­ucts have not been on the mar­ket long enough to exam­ine any longer term effects.

Note that not all com­pa­nies are run­ning ran­dom­ized, con­trolled, stud­ies to show that their prod­uct has a spe­cif­ic impact on brain func­tions. These com­pa­nies, such as Nin­ten­do, base their more lim­it­ed claims on gen­er­al research that shows how men­tal stim­u­la­tion can low­er the prob­a­bil­i­ty of devel­op­ing Alzheimer’s and oth­er demen­ti­as’ symp­toms, via the cog­ni­tive reserve the­o­ry. We can view these untest­ed pro­grams as a new, high-tech, gen­er­a­tion of cross­word puz­zles, that may be use­ful but that can­not make spe­cif­ic brain ben­e­fit claims beyond the gen­er­al “use it or lose it.”

Keep learn­ing by read­ing more arti­cles in the Resources sec­tion, and also please con­sid­er join­ing our free month­ly Brain Fit­ness eNewsletter

This new online resource is based on the con­tent from the book The Sharp­Brains Guide to Brain Fit­ness (May 2009, $19.95), by Alvaro Fer­nan­dez and Dr. Elkhonon Goldberg.

Leave a Reply