Each of our sensory systems are constantly picking up stimulus that exists within our environment. When we become exposed to these things, we begin to make connections and learn about our surroundings and our environment. As a result of this learning process, we begin to form memories, which are described as learned information from a single exposure, or from repetition, of experiences (Gazzaniga, Irvy, & Mangun, 2018). According to research there are major subsets of memory, which are sensory memory, working memory, short-term memory, long-term non declarative memory, and long-term declarative memory. Each of these memory types have been shown to involve various systems within the human brain.
Firstly, sensory memories are described as short-lived sensory information. This information is measurable in milliseconds to seconds, and thus is only available for a short period of time after a person is exposed to it (Gazzaniga, Irvy, & Mangun, 2018). For example, all of the sensory information within our environment, including conversations, road noise, birds chirping, and visual stimuli all form sensory memory traces. That being said, this system of memory has a very high capacity for traces. However, these memory traces will not be encoded and stored unless a person utilizes active memory forming resources to do so (Gazzaniga, Irvy, & Mangun, 2018). There are two major types of sensory memories which are echoic memories, for audible stimulus, and iconic memories for visual stimulus. According to research utilizing mismatch field (MMF), a magnetic way to measure event related potential (ERP), these sensory stimuli form MMF responses at high amplitude during initial exposure, but the response quickly levels off after about 10 seconds (Gazzaniga, Irvy, & Mangun, 2018). In other words, unless given proper attention, we will likely not store most of the sensory information we are exposed to.
The second type of memory is short-term memory. This type of memory is categorized by information that is retained from a timespan of seconds to minutes. This increased time over sensory memory means that this memory system has a much smaller, or more limited, capacity (Gazzaniga, Irvy, & Mangun, 2018). According to popular models, such as the Atkinson and Shiffrin’s modal model, short term memory is a stage within the process of storing new memories. Although this model is highly contested, the modal model suggests that sensory information is sent to short-term memory only when it is given direct attention resources. Only after this occurs will the information be rehearsed and sent to long-term memory (Gazzaniga, Irvy, & Mangun, 2018). Studying those with deficits within their short-term memory capabilities, researchers have found various brain systems involved in the processing of short-term memory. For example, damage to the left perisylvian cortex has shown to impact tasks which rely on short-term memory, and lesions within the inferior parietal cortex have been shown to impact our ability to utilize and form short-term memories (Gazzaniga, Irvy, & Mangun, 2018).
The next type of memory, a subset of short-term memory, is working memory. Working memory describes a limited storge for the retainment of short-term memory and for performing mental operations on the information that is stored in this repository. In other words, we utilize this information for maintenance, or the retainment of information, and for manipulation, or the process of acting upon this information (Gazzaniga, Irvy, & Mangun, 2018). The information found within working memory involves information we are actively utilizing and can include short-term memory, sensory information, and information pulled form long-term memory, such as memorized statistics. According to research, the typical working memory is limited to seven items (Gazzaniga, Irvy, & Mangun, 2018). Based on various studies, researchers believe this type of information is supported via a group of brain areas that support memory systems. For example, patients with damage to the left supramarginal gyrus experience issues with acoustic working information, while damage to the parieto-occipital region of either hemisphere can impact visuospatial short-term memory (Gazzaniga, Irvy, & Mangun, 2018).
Lastly, there is long-term memory, which is information that is retained for long periods of time capable of spanning years (Gazzaniga, Irvy, & Mangun, 2018). This type of memory is divided into subcategories, which are declarative (implicit) and nondeclarative (explicit) memory. Firstly, long-term declarative memory involves episodic memories, which are memories comprised of events that a person has personally experienced, and semantic memory, which is factual knowledge that a person has learned throughout their life. The key to these memory types is the fact that a person is consciously aware of them and able to recall them at will. For example, an episodic memory could involve the who, what, when, and where of a life memory, while semantic memories involve facts you’ve learned such as various state birds (Gazzaniga, Irvy, & Mangun, 2018). Utilizing cases such as those presented by patient E.E. who had the majority of his medial temporal lobe removed, researchers have found that the medial temporal lobe is essential for the capacity of declarative memories (Gazzaniga, Irvy, & Mangun, 2018).
The second type of long-term memory are those referred to as non-declarative memories. As the name implies, these memories differ from declarative memories as they cannot be verbally expressed or declared. Instead, these memories are demonstrated through performance (Gazzaniga, Irvy, & Mangun, 2018). Non-declarative memories can be further broken down into several examples such as procedural memory and priming. Procedural memory, supported by the basal ganglia, describes memories that are required for tasks such as motor skills, like learning to drive manual car, and cognitive skills, like learning math equations (Gazzaniga, Irvy, & Mangun, 2018). Priming refers to a person’s ability to change their response to specific stimuli after they have been previously exposed to it. Priming is supported by the core semantic network, which involves the anterior temporal lobe, superior temporal sulcus, and the ventral prefrontal cortex (Gazzaniga, Irvy, & Mangun, 2018).