At some point in the future, cognitive neuroscience will be able to describe the algorithms that drive structural neural elements into the physiological activity that results in perception, cognition, and perhaps even consciousness. To reach this goal, the field has departed from the more limited aims of neuropsychology and basic neuroscience. Simple descriptions of clinical disorders are a beginning, as is understanding basic mechanisms of neural action. The future of the field, however, is in working toward a science that truly relates brain and cognition in a mechanistic way.”
— M. Gazzaniga

What is Cognitive Neuroscience?

Cognitive neuroscience is a field of neuroscience that seeks to bridge the gap between biological sciences and behavioral sciences by exploring the processes that contribute to complex mental functions.

The Roles of Artificial Intelligence in Neuroscience

Artificial intelligence is able to benefit from neuroscience and neuroscience is able to benefit from AI. The goal of cognitive neuroscience is to understand human behaviors, but there are many challenges in this area of research including efficiency and researcher bias. Romy Lorenz, a cognitive neuroscientist, worked to develop what she calls an “AI neuroscientist” during her PhD work. The AI neuroscientist is able to analyze brain data in real time while participants are in a brain scanner which makes the brain scanning more powerful and flexible. The AI neuroscientist may allow neuroscientists to analyze multiple behaviors at once and may be key to decoding other human behaviors. Some other benefits to AI in neuroscience research is the ability to erase human subjectivity from research and make calculated predictions about behavior.

Highfield, Roger. “This AI Could Hold the Key to Decoding Human Intelligence.” WIRED, WIRED UK, 11 Sept. 2017, www.wired.co.uk/article/automatic-neuroscientist-ai-brain-experiments. 

The Roles of Neuroscience in Artificial Intelligence

Neuroscience can be useful in improving and developing current and new forms of AI. The mapping of neural networks, especially deep learning networks, can be inspiration for new algorithms for AI programs. Neural networks can also contribute to advances in AI reasoning and a new understanding how these programs answer complex problems. For example, by using models similar to synaptic tuning, forms of AI are able to reason through difficult tasks by tuning inputs for accuracy and efficiency. Neuroscience also compliments many of the foundational practices of AI such as logic and mathematics approaches with biological computations. The two disciplines end up becoming quite complimentary to each other making the relationship between neuroscience and AI one that could generate answers to many seemingly impossible questions.

Lee, Justin https://medium.com/swlh/how-neuroscience-enables-better-artificial-intelligence-design-5d254098470b

Deep Learning

Some forms of artificial intelligence is modeled after neural networks. One very important model for AI is deep learning. Deep learning refers to machine learning technique that uses deep (also called layered) neural networks that are modeled after the human brain (Waldrop). These deep learning networks are powerful predicative models for many aspects of behavior, neuronal activity, and cortical activity making it a powerful tool for neuroscientists (Storrs and Kriegeskorte 2019). Deep learning may also drive development of new models of AI that are better able to learn more conceptual materials and excel at performances of real-world tasks. This could lead to advances in AI that could greatly benefit our daily lives. So far, deep learning networks enable technology features behind speech recognition (like Siri or Alexa), image recognition and self-driving cars.

The ultimate competitive advantage is being cognitive.

—Ginni Rometty

                         
M. Mitchell Waldrop. News Feature: What are the limits of deep learning? Proceedings of the National Academy of Sciences Jan 2019, 116 (4) 1074-1077; DOI: 10.1073/pnas.1821594116     
        
Storrs K.R. Kriegeskorte N. Deep learning for cognitive neuroscience.
in: Gazzaniga M. The Cognitive Neurosciences. 6th Edition.  MIT Press,         Boston; 2019