Most of discoveries made by the human kind came through combination of insight, intuition and imagination. First we perceive the reality that comes through our senses, and build an internal representation or model of a given aspect of the reality. Then from that original model, we continue creating variations, adding more features, and then we arrive to a more complex and complete model. The process whereby we starts from perceptions to later create concepts and models has been a matter a philosophical analysis. Kant said that kind of knowledge is “a posteriori”.
The first cognitive models, were more an idealistic representation rather than an cartesian model created through the scientific method or laboratory experiments. It was later on, with the development of the cognitive science and psychology that a new set of cognitive models arrived as a result of laboratory experiments. Actually the models in cognitive science can be categorized as verbal-conceptual, mathematical, or computational. Computational models describe the cognitive process using algorithms, mathematical models using mathematical equations, and verbal-conceptual describes entities, the relation and process just using declarative sentences.
The last years has had a steadily development of cognitive models, cognitive neuroscience , paradigms, cognitive architectures, software, tools, and applications that allow us to have a more detailed and accurate representation of human cognitive process. What is better: the connectionist model of cognition? Or the declarative logic based cognitive modeling? It depends on the specific purpose and application. The number of options to model the cognitive process are overwhelming. Eventhough there are several cognitive architectures: SOAR, EPIC, ACT-R, CLARION, they are still constrained in terms of the functionality they can provide, prediction strenght and theory support.
A number of application domains result from the computational psychology, going from a better understanding of our cognitive and mental process, to the development of applications, systems or tools to treat mental health illness and enhancement of our cognitive process, social and organizational simulation, and interpretation of psychological data, artificial intelligence and affective computing. We expect the cognitive models will approach and converge slowly to a more accurate representation of our mind. And the direct and side benefits from that are still unpredictable, but certainly encouraging.
 O’Reilly, Randall & Munakata, Yuko. (2000). Computational Explorations in Cognitive Neuroscience Understanding the Mind by Simulating the Brain. 10.7551/mitpress/2014.001.0001. https://grey.colorado.edu/CompCogNeuro/index.php/CECN
 Sun, R. (Ed.). (2008). The Cambridge handbook of computational psychology. Cambridge University Press. https://doi.org/10.1017/CBO9780511816772
Photo by Rodrigo Acuña for Psycognet