Language and cognition researchers confirm how the brain is wired to learn concepts—and test a way to use electricity to improve dementia treatments. Think about the last time you looked at a dog. Whether or not you noticed it, you probably thought the word dog. Now think about the last time you heard a dog barking. In your mind’s eye, you likely saw an image of a dog. And when you've read or heard the word dog, again, chances are good that images or sounds of some kind of dog flashed through your imagination.
This is the semantic memory at work—the part of your memory that manages ideas and concepts that aren’t linked to specific personal experiences. Intuitively, we know this is how concepts are learned and held in the brain.
Now new research from Temple's Memory, Cognition, and Concepts Laboratory lends empirical evidence to our hunches about how the brain acquires and represents concepts. This work may also suggest new avenues for the treatment of debilitating language loss associated with Alzheimer’s disease and other forms of dementia.
In the study – funded by the NIH National Institute on Deafness and Other Communication Disorders and published May 16 in the journal Frontiers in Human Neuroscience – Assistant Professor and lead author Jinyi Hung and Associate Professor Jamie Reilly (both in the College of Public Health's Department of Communication Sciences and Disorders) collaborated with colleagues at the University of Pennsylvania used functional magnetic resonance imaging (fMRI) to map regions of the brain as they responded to visual and auditory representations of simple concepts. They also tracked how each of these regions interacted with one another. Their findings offer important takeaways that shed light on how the brain retains concepts and possible ways to improve treatment for patients with dementia.
Concepts Need A Connected Brain
The results suggest that there are core regions of the brain that bind and integrate different components of knowledge (for example, the color or sound of an object) and that these “hub” regions interact with other parts of the brain that are active during perception of the same objects.
One big takeaway of the research, Reilly says: "There is no single part of the brain that represents concepts. Our object knowledge is rich, multimodal, and spread out over a variety of areas. We do know, however, that certain parts of the brain that act as unifying hubs are especially important for representing knowledge." One of those parts is the anterior temporal lobes (or ATL for short). This is one part of the brain where features such as sound and vision are likely to converge. We also know from post-mortem studies of dementia that this brain region is especially vulnerable to neuropathologies such as Alzheimer’s disease and frontotemporal dementia. “So the information we’ve gleaned from healthy young adults yokes with what we know about a range of neurodegenerative conditions,” Reilly says.
Namely, that semantic memory relies on many inputs, and when those connections begin to fade, so do the concepts—and the language that expresses them.
A Link Between Electricity and Language Therapy
The researchers understood that if they could identify where those hubs are and how they work, they might be helpful in treating dysfunction in semantic memory. Reilly, who specializes in the aging brain, is especially interested in the possibility of directly stimulating one of these areas using low-voltage electrical currents with the goal of boosting language functioning. So the group performed a companion study of patients suffering from primary progressive aphasia (PPA) concomitant with Alzheimer’s disease or other dementia. The researchers wanted to determine how transcranial direct current stimulation (or tDCS) would affect language therapy used with these patients. That tDCS would be a supplement to, not a replacement for, behavioral language therapy is important to note.
One feature common in PPA is progressive naming impairment. Patients start forgetting words, and they continue to lose words in ways that are nearly impossible to predict. Currently, there’s no way to tell what words will be lost next and when. Treatment is difficult because there’s no extant model from similar-looking conditions that are effective in treating PPA. For example, language disorders associated with brain injuries such as stroke are treated with certain protocols. In PPA, there is only decline; the best we can hope for, with what we know about the disease so far, is to try to stave off progression and maintain those words which the patient already has.
To further complicate matters, three variants of PPA have been observed, each of which shows up differently from individual to individual. Because there’s no “one-size-fits-all” intervention, a behavioral component to language therapy is essential.
For this study the researchers looked particularly at semantic variant PPA (svPPA), which is characterized by profound impairments in word and object knowledge; such impairments are associated with degeneration of parts of the ATL. Using Reilly’s treatment protocol for svPPA-related semantic anomia, each patient worked with a short list of words from the semantic categories most important for everyday living – things like familiar people, foods, hygiene, and household items.
Some words were used for training, others as controls. In each therapy session, study volunteers received tDCS stimulation, then began behavioral treatment which continued for 5 to 10 minutes after electrical stimulation had ceased. The results of trained items compared with untrained items showed a promising pattern. The baseline naming accuracy before the test sessions was 69 percent for trained items and 40 percent for untrained items.
After two weeks, accuracy climbed to 77 percent for trained items, while untrained items remained mostly level, at 41 percent. At the six-month follow-up, accuracy on trained items dipped slightly, to 63 percent, and untrained items dropped precipitously, to 26 percent. This suggests that even if behavioral training combined with tDCS doesn’t reverse PPA, it may slow its functionally debilitating symptom of language loss.
Until There's a Cure, There's Improved Treatment
In light of the aging population and generally longer life spans, dementia is expected to be a critical issue with immense impacts. Accordingly, advances in early detection and pharmaceutical treatments are helping people manage disease progression better. But, as the authors discuss, techniques like these, while not a cure, can promote functional independence, ease the emotional and social impacts of dementia, and delay the need for institutionalized care.
“Dementia impacts so many of our loved ones,” Reilly says, “it is crucial that we develop interventions that will prolong functional independence and communication for the millions affected by dementia. Outright prevention at the cellular/molecular level in the form of a cure or vaccine is the ultimate hope in biomedical research. However, we cannot stand idle while awaiting a cure. To this end, our behavioral treatment offers a proactive and effective approach to maintaining elements of language (and cognition) as dementia progresses. We are excited to continue this truly important work.”