Para distinguir contextos, os animais pensam probabilisticamente, o estudo sugere
![](data:image/jpeg;base64,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)
Um labirinto no laboratório Wilson do MIT. Um roedor deve inferir que este éum contexto diferente do que, digamos, um labirinto que utilizou diferentes sinais de forma ou um que tivesse um braa§o adicional. Um novo estudo sugere que eles pesem as probabilidades de fazaª-lo. Crédito: Peter Goldberg
Entre as muitas coisas que os roedores ensinaram aos neurocientistas, estãoo fato de que em uma regia£o chamada hipocampo, o cérebro cria um novo mapa para cada contexto espacial aºnico - por exemplo, uma sala ou labirinto diferente. Mas os cientistas atéagora tem se esforçado para aprender como os animais decidem quando um contexto énovo o suficiente para merecer a criação, ou pelo menos a revisão, desses mapas mentais. Em um estudo no eLife , os pesquisadores do MIT e Harvard propaµem um novo entendimento: o processo de "remapeamento" pode ser modelado matematicamente como uma faznha do raciocanio probabilastico dos roedores.
A abordagem oferece aos cientistas uma nova maneira de interpretar muitos experimentos que dependem da medição do remapeamento para investigar a aprendizagem e a memória. O remapeamento éparte integrante dessa busca, porque os animais (e as pessoas) associam o aprendizado ao contexto , e os mapas do hipocampo indicam em que contexto o animal acredita estar.
"As pessoas perguntaram anteriormente 'quemudanças no ambiente fazem com que o hipocampo crie um novo mapa?' mas não houve respostas claras ", disse o principal autor Honi Sanders. "Depende de todos os tipos de fatores, o que significa que a forma como os animais definem o contexto foi envolta em mistanãrio".
Sanders éum pa³s-doutorado no laboratório do co-autor Matthew Wilson, Sherman Fairchild Professor no Instituto Picower de Aprendizagem e Mema³ria e nos departamentos de Biologia e Ciências Cognitivas e do Canãrebro do MIT. Ele também émembro do Centro de Canãrebros, Mentes e Ma¡quinas. A dupla colaborou com Samuel Gershman, professor de psicologia em Harvard no estudo.
Fundamentalmente, um problema com o remapeamento que frequentemente levou os laboratórios a relatar resultados conflitantes, confusos ou surpreendentes éque os cientistas não podem simplesmente garantir a seus ratos que eles passaram do Contexto A experimental para o Contexto B, ou que ainda estãono Contexto A, mesmo se alguma condição ambiental, como temperatura ou odor, tiver sido alterada inadvertidamente. Cabe ao rato explorar e inferir que condições como a forma do labirinto, ou cheiro, ou iluminação, ou a posição de obsta¡culos e recompensas, ou a tarefa que eles devem executar, mudaram ou não o suficiente para provocar uma remapeamento parcial.
Sanders, Wilson e Gershman argumentam que, em vez de tentar entender as medidas de remapeamento com base no que o projeto experimental deve induzir, Sanders, Wilson e Gershman argumentam que os cientistas devem prever o remapeamento, matematicamente, contabilizando o raciocanio do rato usando estatasticas bayesianas, que quantificam o processo de comea§ar com um incerto. pressuposto e depois atualiza¡-lo a medida que novas informações surgem.
"Vocaª nunca experimenta exatamente a mesma situação duas vezes. A segunda vez ésempre um pouco diferente", disse Sanders. "Vocaª precisa responder a pergunta: 'Essa diferença éapenas o resultado da variação normal nesse contexto ou essa diferença érealmente um contexto diferente?' Na primeira vez em que vocêexperimenta a diferença, não pode ter certeza, mas depois de experimentar o contexto várias vezes e ter uma noção de qual variação énormal e qual não anã, vocêpode perceber imediatamente quando algo estãofora de linha . "
Â
O trio chama sua abordagem de "inferaªncia do estado oculto" porque, para o animal, a possível mudança de contexto éum estado oculto que deve ser inferido.
No estudo, os autores descrevem vários casos em que a inferaªncia de estado oculto pode ajudar a explicar o remapeamento, ou a falta dele, observado em estudos anteriores.
Por exemplo, em muitos estudos, tem sido difacil prever como a alteração de algumas dicas pelas quais um roedor navega em um labirinto (por exemplo, uma luz ou uma campainha) influenciara¡ se ele faz um mapa completamente novo ou remapeia parcialmente o atual e como Muito de. Principalmente, os dados mostraram que não háuma relação a³bvia "individual" entre mudança de sugestãoe remapeamento. Mas o novo modelo prevaª como, a medida que mais pistas mudam, um roedor pode passar de uma incerteza sobre se um ambiente énovo (e, portanto, parcialmente remapear) para ter certeza suficiente disso para remapear completamente.
Em outro, o modelo oferece uma nova previsão para resolver uma ambiguidade de remapeamento que surgiu quando os cientistas "incrementaram" a forma dos gabinetes de roedores. Va¡rios laboratórios, por exemplo, encontraram resultados diferentes quando familiarizaram os ratos com ambientes quadrados e redondos e depois tentaram medir como e se remapearam quando colocados em formas intermedia¡rias, como um octa³gono. Alguns laboratórios viram remapeamento completo, enquanto outros observaram apenas remapeamento parcial. O novo modelo prevaª como isso pode ser verdade: os ratos expostos ao ambiente intermediário após um treinamento mais longo seriam mais propensos a remapear completamente do que aqueles expostos a forma intermedia¡ria no inicio do treinamento, porque com mais experiência, eles teriam mais certeza de seus ambientes originais e, portanto, mais certo de que o intermediário foi uma mudança real.
A matemática do modelo inclui atéuma varia¡vel que pode explicar diferenças entre animais individuais. Sanders estãoanalisando se repensar resultados antigos dessa maneira pode permitir que os pesquisadores entendam por que roedores diferentes respondem tão variavelmente a experimentos semelhantes.
Por fim, disse Sanders, ele espera que o estudo ajude os pesquisadores de remapeamento a adotar uma nova maneira de pensar em resultados surpreendentes - considerando o desafio que seus experimentos representam para os sujeitos.
"Os animais não tem acesso direto a s identidades de contexto, mas precisam inferi-las", disse ele. "As abordagens probabilasticas capturam a maneira como a incerteza desempenha um papel quando ocorre a inferaªncia. Se caracterizarmos corretamente o problema que o animal estãoenfrentando, podemos entender resultados diferentes em situações diferentes, porque as diferenças devem resultar de uma causa comum: a maneira como oculta inferaªncia de estado funciona ".