This system is a form of a computational model that helps us to enlighten ourselves about the complex characteristics of human memory functioning. As the name suggests, the whole system of storing and retrieving memory is based on several computational elements, performing their desired tasks simultaneously or in the required fashion to carry out the day-to-day working of our memory. These elements establish a connection between themselves, and the strength of these connections governs the retrieval or long-lasting effect of memory. The stronger the bond, the longer the memory will stay in mind, and correspondingly the remembrance of such memories will be better compared to those in which the strength of the connection is weak. This process of connections is a continual one.
This model is generally referred to as the PDP model. The model consists of computational elements, also known as units, which are neurally inspired. It assumes that the information is placed in the brain using different activation patterns. Each unit of the PDP model can take a value of activation between minimum and maximum (0 and 1). These activations are propagated among different units in the simple process of thinking about something. When the connection is positive, it increases the amount of activation.
On the contrary, when the connection is negative, it reduces the amount of activation between various units. In systems like this, knowledge is present between the connections established between units, determining what pattern will emerge from the representation of input or data. The firmness of such connections governs the act of recollection. For example, suppose one is standing between two groups of people, continuously chanting about any topic. In that case, this person will only be able to pick up some recollections from both groups simultaneously.
Unlike parallel, in a serial type of processing, the phenomenon involves the sequential filtering flow without any overlap among the different processes. In the case of serial purifying, the elements are searched one after the other in a serial order to find the target element. This approach comparatively reduces the accuracy of the search, and the time to do so is also increased when dealing with a larger number of objects.
It includes
The same underlying principle governs all cognitive behaviors. Units adjust their current activations based on the total inputs received via connections between units. Different models adopt different policies for the aggregation of inputs and adjustment of activations. Cognitive filtering involves the propagation of activations between units via connections established in neural networks.
The same underlying principle governs all cognitive behaviors. Units adjust their current activations based on the total inputs received via connections between units. Different models adopt different policies for the aggregation of inputs and adjustment of activations. Cognitive filtering involves the propagation of activations between units via connections established in neural networks.
PDP models are dynamic. The transfer of activations is not uni-directional; instead, it is bi-directional. This means that when a signal is transmitted from unit A to unit B, it will receive a signal back from unit B, which makes the processing interactive. As a result, the unit alters the data, changes the input details, and transfers it downstream. This phenomenon happens many times and imparts a dynamic nature to the model.
The data that is the base of the network and allows it to behave properly is not stored in the form of separate data structures in a separate store. However, they are directly encoded in the connections of these networks to generate the inter representations and output desired at the activity time.
Processing, learning, and representations are continuous in the PDP structure. Because representations are coded as distributed patterns of activations among units, these representations of different items can be similar to one another. These similarities provide the basic mechanism for generalizing activations in the network substructure. It was concluded that similar patterns of activations would produce familiar representations and outputs of the same kind.
The statistical structure of the environment responsible for the creation of responses is critically important in understanding the processing model. Learning is the result of patterns of activations occurring within neural networks, the errors in such and violated expectations are generated in the daily experiences. The PDP phenomenon depends on the statistical structure present in everyday structures.
In the PDP framework, the stimulus representations in the consciousness correspond to the patterns of activations generated over all units involved. Any input is assumed to be represented as patterns of activations distributed among many neurons distributed all over the mind. Each neuron is supposed to participate in the representations of many different inputs. These representations are related to almost every cognitive content, such as colors, pictures, words, letters, structures, etc., all of which are thought to take the form of patterns among widely distributed neurons in the system.
We have had an overview of the whole framework of PDP. In conjunction with other systems, these models are an initiative and tend to provide solutions to various challenges in the neuroscience or applied psychology sciences. This framework has offered new productive ways of learning in cognitive sciences and innovative ways of dealing with challenges still unmastered by psychologists. It has also enabled us to address some computational challenges that motivated the search to find alternative symbolic approaches. In the world of technological advancements, one such is Artificial intelligence. These PDP volumes play a vital role in machine learning, which is an outgrowth or extension of Artificial intelligence. In the research field related to long-term memory, the success and failures both of the Parallel Distributed Processing model have contributed much to enable a better understanding of the same. It is important to understand that these models have advantages and disadvantages, and before using such structures, they should be considered.
Physics
Biology
Chemistry
Mathematics
Economics
Psychology
Environmental Science
Social Studies
Fashion Studies
English