"Long before it's in the papers"
January 27, 2015


Human reasoning attributed to special brain network

Dec. 3, 2014
Courtesy of UC Berkeley
and World Science staff

When it comes to get­ting out of a tricky situa­t­ion, we hu­mans have an edge over our evo­lu­tion­ary relatives. Take, for ex­am­ple, the Apol­lo 13 voy­age in which en­gi­neers, against all odds, im­pro­vised a chem­i­cal fil­ter on a lu­nar mod­ule to pre­vent car­bon di­ox­ide build­up from kill­ing the crew.

Sci­en­tists say they have found mount­ing ev­i­dence that helps ex­plain how hu­mans have ex­celled at “rela­t­ional rea­son­ing,” a skill in which we dis­cern pat­terns and rela­t­ion­ships to make sense of seem­ingly un­re­lat­ed in­forma­t­ion, such as solv­ing prob­lems in un­fa­mil­iar cir­cum­stances.

Their find­ings link such reasoning to sub­tle changes in brain ar­eas known the front­al and pa­ri­e­tal lobes, at the top of the head and behind the forehead. Among oth­er things, the sci­en­tists say the net­work of nerve cell con­nec­tions in this re­gion—the “frontopa­ri­e­tal net­work”—aids in anal­y­sis, mem­o­ry re­triev­al, ab­stract think­ing and problem-solv­ing, and has the flu­id­ity to adapt ac­cord­ing to the task at hand.

“This re­search has led us to take se­ri­ously the pos­si­bil­ity that tweaks to this net­work over an ev­o­lu­tion­ary timescale could help to ex­plain dif­fer­ences in the way that hu­mans and oth­er pri­ma­tes solve prob­lems,” said Uni­vers­ity of Cal­i­for­nia Berke­ley neu­ro­sci­ent­ist Sil­via Bunge, the stu­dy’s prin­ci­pal in­ves­ti­ga­tor.

“It’s not just that we hu­mans have lan­guage at our dis­pos­al. We al­so have the ca­pa­city to com­pare and in­te­grate sev­er­al pieces of in­forma­t­ion in a way that oth­er pri­ma­tes don’t,” she added.

In re­view­ing doz­ens of stud­ies – in­clud­ing their own – that use brain im­ag­ing, neu­ropsy­chol­ogy, de­vel­op­men­tal cog­ni­tive and oth­er meth­ods, the re­search­ers con­clud­ed that anatom­i­cal changes in the lat­er­al frontopa­ri­e­tal net­work over mil­len­nia have boosted our rea­son­ing.

There is “sup­port­ing ev­i­dence across species,” added Mi­chael Vendetti, co-author of the study and a post­doc­tor­al re­searcher in neu­ro­sci­ence at UC Berke­ley.

In rela­t­ional rea­son­ing, we make com­par­isons and find equiv­a­len­cies, as one does in al­ge­bra, for ex­am­ple. “First-order” com­par­isons iden­ti­fy sev­er­al types of rela­t­ion­ship be­tween two items or ac­ti­vi­ties: se­mantic (ham­mer is used to hit a nail); nu­mer­ic (four is great­er than two); tem­po­ral (we get out of bed be­fore we go to work) or vi­su­ospa­tial (the bird is on top of the house). “Second-order” or higher-order com­par­isons take this a step fur­ther by equat­ing two or more sets of first-order rela­t­ions (a chain is to a link as a bou­quet is to a flow­er).

The re­search­ers iden­ti­fied three brain ar­eas key to rela­t­ional rea­son­ing, called the rostrolat­er­al prefront­al cor­tex, the dorsolat­er­al prefront­al cor­tex and the in­fe­ri­or pa­ri­e­tal lob­ule, with the rostrolat­er­al re­gion more ac­tively en­gaged in second-order rela­t­ional rea­son­ing.

Al­so cru­cial to their find­ing was a study led by Ox­ford Uni­vers­ity neu­ro­sci­ent­ist Mat­thew Rush­worth that com­pared neu­ral pat­terns in hu­mans and ma­caque mon­keys. While hu­mans, apes and mon­keys found to share si­m­i­lar­i­ties in the front­al and pa­ri­e­tal brain re­gions, ac­ti­vity in the hu­man rostrolat­er­al prefront­al cor­tex dif­fered sig­nif­i­cantly from that of the ma­caque mon­key’s front­al cor­tex, the study found.

“We had hy­poth­e­sized that there could have been ev­o­lu­tion­ary changes to this re­gion to sup­port our rea­son­ing abil­ity, so we were really ex­cit­ed when Rush­worth and his col­leagues came out with these find­ings,” Vendetti said.

Mean­while, in the be­hav­ior­al stud­ies they an­alyzed, hu­mans were found to use higher-order strate­gies to guide their judg­ment while non-hu­man pri­ma­tes re­lied more heavily on per­cep­tu­al si­m­i­lar­i­ties and were slower at rea­son­ing and problem-solv­ing.

“These re­sults do not nec­es­sarily prove that non-hu­man pri­ma­tes are un­able to rea­son us­ing higher-order think­ing, but if it is pos­si­ble to train non-hu­mans to pro­duce hu­man-like per­for­mance on tasks as­so­ci­at­ed with higher-order rela­t­ional think­ing, it is cer­tainly not some­thing that comes nat­u­rally to them,” the study con­clud­ed. The work is pub­lished in the Dec. 3 is­sue of the jour­nal Neu­ron.

* * *

Send us a comment on this story, or send it to a friend

Sign up for


On Home Page         


  • St­ar found to have lit­tle plan­ets over twice as old as our own

  • “Kind­ness curricu­lum” may bo­ost suc­cess in pre­schoolers


  • Smart­er mice with a “hum­anized” gene?

  • Was black­mail essen­tial for marr­iage to evolve?

  • Plu­to has even cold­er “twin” of sim­ilar size, studies find

  • Could simple an­ger have taught people to coop­erate?


  • F­rog said to de­scribe its home through song

  • Even r­ats will lend a help­ing paw: study

  • D­rug may undo aging-assoc­iated brain changes in ani­mals

When it comes to getting out of a tricky situation, we humans have an evolutionary edge over other primates. Take, for example, the Apollo 13 voyage in which engineers, against all odds, improvised a chemical filter on a lunar module to prevent carbon dioxide buildup from killing the crew. Scientists say they have found mounting evidence that helps explain how humans have excelled at “relational reasoning,” a skill in which we discern patterns and relationships to make sense of seemingly unrelated information, such as solving problems in unfamiliar circumstances. Their findings, reported in the Dec. 3 issue of the journal Neuron, suggest that subtle changes in brain areas known the frontal and parietal lobes are linked to superior cognition. Among other things, the scientists say the network of nerve cell connections in this region—the “frontoparietal network”—aids in analysis, memory retrieval, abstract thinking and problem-solving, and has the fluidity to adapt according to the task at hand. “This research has led us to take seriously the possibility that tweaks to this network over an evolutionary timescale could help to explain differences in the way that humans and other primates solve problems,” said University of California Berkeley neuroscientist Silvia Bunge, the study’s principal investigator. “It’s not just that we humans have language at our disposal. We also have the capacity to compare and integrate several pieces of information in a way that other primates don’t,” she added. In reviewing dozens of studies – including their own – that use brain imaging, neuropsychology, developmental cognitive and other investigative methods, Bunge and fellow researchers concluded that anatomical changes in the lateral frontoparietal network over millennia have boosted our reasoning. There is “supporting evidence across species,” added Michael Vendetti, co-author of the study and a postdoctoral researcher in neuroscience at UC Berkeley. In relational reasoning, we make comparisons and find equivalencies, as one does in algebra, for example. “First-order” comparisons identify several types of relationship between two items or activities: semantic (hammer is used to hit a nail); numeric (four is greater than two); temporal (we get out of bed before we go to work) or visuospatial (the bird is on top of the house). “Second-order” or higher-order comparisons take this a step further by equating two or more sets of first-order relations (a chain is to a link as a bouquet is to a flower). The researchers identified three brain areas key to relational reasoning, called the rostrolateral prefrontal cortex, the dorsolateral prefrontal cortex and the inferior parietal lobule, with the rostrolateral region more actively engaged in second-order relational reasoning. Also crucial to their finding was a study led by Oxford University neuroscientist Matthew Rushworth that compared neural patterns in humans and macaque monkeys. While humans, apes and monkeys found to share similarities in the frontal and parietal brain regions, activity in the human rostrolateral prefrontal cortex differed significantly from that of the macaque monkey’s frontal cortex, the study found. “We had hypothesized that there could have been evolutionary changes to this region to support our reasoning ability, so we were really excited when Rushworth and his colleagues came out with these findings,” Vendetti said. Meanwhile, in the behavioral studies they analyzed, humans were found to use higher-order strategies to guide their judgment while non-human primates relied more heavily on perceptual similarities and were slower at reasoning and problem-solving. “These results do not necessarily prove that non-human primates are unable to reason using higher-order thinking, but if it is possible to train non-humans to produce human-like performance on tasks associated with higher-order relational thinking, it is certainly not something that comes naturally to them,” the study concluded.