Forms of memory encoding
Verbal codes, images, propositions and productions
God gave us memories that we might have roses in December. ~J.M. Barrie, Courage, 1922
Working memory refers to the temporary holding space for information that is being manipulated and processed in order to perform such functions as comprehension, decision-making and problem solving. In contrast, long-term memory is the relatively permanent part of memory that is capable of storing large amounts of information for a long period of time. Encoding, which can occur either automatically or as the result of concerted effort, refers to the way in which the information is processed for storage as it moves from working to long-term memory (Ormrod, 2004; Santrock, 2003). The encoding process basically consists of the acquisition of information and the subsequent initial formation of a memory trace. It is essentially the preparation of information for storage in long term memory and is accomplished by making new information meaningful and/or integrating it with known information.
Typically, information that is stored only temporarily in working memory is encoded primarily in acoustic form. In contrast, most information stored in long-term memory is encoded primarily in semantic code – i.e., with reference to the meaning of the words rather than merely the sounds (Baddeley, 1996: Conrad, 1964; Grossman & Eagle, 1970). This semantic coding can take various forms including, verbal codes, imagery, propositions and productions (Ormrod, 2004).
Verbal coding refers to the use of actual words as symbols for the objects they represent. This form of coding is evidenced by that fact that we have linguistic labels for the entire spectrum of human experience, from objects to experiences. We understand “pen,”, “picnic,”, “birthday,”, “accident,” and “love” because these words have been encoded in memory as having a specific meaning. Whereas these verbal codes can certainly generate different specific examples in different individuals (i.e., a Mont Blanc fountain pen versus a standard Bic ballpoint), the general category is a well understood concept easily recognizable based on the information stored in memory associated with the word.
Psychological studies have shown that the strength and duration of a memory is significantly influenced by the cognitive operations engaged during initial encoding of that experience (Carlson, Martin & Buskist, 2004; Ormond, 2004). It is generally accepted that meaningful words – i.e., those with a high frequency of occurrence in language and print, those easily pronounced and those that have many familiar associations to common item – are more easily encoded and thus more easily retained and recalled. Additionally, concrete words – those that refer to actual physical objects – as opposed to abstract words – those that refer to more conceptual and intangible ideas – have a greater number of associations, thus facilitating better encoding (Terry, 2000).
Although somewhat controversial and difficult to isolate, most psychologists posit the existence of mental imagery in the form of visual, auditory or olfactory images, as a distinct process of memory encoding (Carlson, Martin & Buskist, 2004; Ormrod, 2004). These memories exist as recollections of such things as a familiar face, a musical score, or the scent of a favorite perfume. Researchers studying this type of memory have the unique challenge of separating the processes of encoding imagery versus encoding verbal representations. Paivio (1971) proposed that the storage of verbal and visual knowledge was interconnected but functionally independent. He argued that concrete objects tended to be stored as images, while abstract concepts were stored in verbalized code. Additional research has indicated that areas in the left hemisphere are typically more active during the study and recall of verbal items while areas of the right hemisphere are more active during the encoding of visual or photograph scenes (Brewer, Zhao, Desmond, Glover, & Gabrieli, 1998; Wagner et. al., 1998).
Imagery is frequently used to encode information for transfer to long-term memory. Some types of information are more amenable to imagery encoding than others. The dual-coding theory, proposed by Paivio (1986; Clark & Paivio, 199) attempts to account for why concrete information is typically remembered better than abstract information. For example, in a serial-learning task, subjects remember words like turkey, piano, or car better than equally frequent but abstract words like freedom, idea, or happiness. The same effect has been demonstrated with sentences. Paivio explained that concrete words have a readily available imagery code that abstract words lack. Therefore, abstract words must be encoded as the words themselves, or perhaps as abstract propositions. Concrete words may be encoded in these ways as well, but they also have an additional very useful memory code available through visual imagery. This accounts for their greater memorability. Although abstract words could conceivably be encoded by visual imagery, and they likely are on occasion, it is a considerably less efficient code for abstract than for concrete words.
Among the earliest researchers to engage in the serious consideration of the role of visual imagery was Sir Francis Galton (1822-1911) in his studies of individual differences. In his research Galton asked his subjects to describe and rate their visual images. His conclusions depicted a wide range of clarity varying from vivid mental images to the relative absence of clear visual recollections among his test subjects (Gleitman, 2003).
Modern studies can utilize technology to produce compound sinusoidal gratings (plaid patterns) or computer-generated, synthetic human faces as memory test items. These represent stimuli for which verbal association is relatively difficult, thus indicating a capacity for image encoding (Sekular & Kahana, n.d.). These studies, however, tend to represent rather rote memorization of visual stimuli, thus disregarding the widely accepted notion of the importance of meaning and emotion in the processing and encoding of information.
Propositions refer to “the smallest unit of information that can stand alone as an assertion and that can be judged as true or false” (Schunk, 2004, p. 159). Additionally, each proposition can be broken down in component parts including arguments – the objects or events- and relations – the descriptions of or associations between the arguments (Ormrod, 2004). As such, they represent basic units of knowledge and meaning, combining in long-term memory to form associative network structures which yield a comprehensive understanding of a subject.
From the sentence “Tom’s tiny dog bit the mailman.” there are three simpler units of meaning that can be derived: (1) Tom has a dog; (2) The dog is tiny; (3) The dog bit the mailman. Each of these simpler assertions represents a proposition that can be judged true or false, and together they combine to form a more complex idea. Although the exact nature and operation of propositions remains largely unknown, propositions are theorized as existing in a hierarchical organization and represent the model of how chunks of meaning are stored in human memory (Ormrod, 2004; Schunk, 2003).
Productions refer to “networks of condition-action sequences and rules in which the condition is a set of circumstances that activates the system and the action is the set of activities that occurs as a result” (Schunk, 2004, p. 170). Procedural knowledge is encoded in sequences of productions, referred to as production systems, that can be represented as if-then statements characterized by a recognize-act series of events (Ormrod, 2004; Schunk, 2003).
One set of production sequences for encountering a traffic light might include:
• IF the light is red, THEN apply brake
• IF the light turns green, THEN release brake
• IF brake is released, THEN step on gas pedal
As such, productions represent the stored cognitive processing referenced in reaction to environmental cues and events.
Each of these processing modes represents a unique method of memory encoding. However, in the reality of daily living such distinctions are typically not so clear or separate. Reacting to the environment involves a complex, and not fully understood interdependence between various cognitive processing systems.
Baddeley, A. (1996). Short-term memory for word sequences as a function of acoustic, semantic and formal similarity. Quarterly Journal of Experimental Psychology, 18, 362-365.
Brewer, J. B., Zhao, Z., Desmond, J. E. Glover, G. H. & Gabrieli, J. D. E. (1998). Making memories: Brain activity that predicts how well visual experience will be remembered. Science, 281, 1185-1187.
Carlson, N. R., Martin, G. N. & Buskist, W. (2004). Psychology. New York, NY: Pearson Education.
Clark, J. M. & Paivio, A. (1991). Dual coding theory and education. Educational Psychology Review, 3(3), 149-170.
Conrad, R. (1964). Acoustic confusions in immediate memory. In R. J. Sternberg (2003). Cognitive psychology, (3rded.). Belmont CA: Wadsworth.
Gletiman, H. (2003). Psychology (6th ed.). New York, NY: W. W. Norton & Company.
Ormrod, J. E. (2004). Human learning (4th ed.). Upper Saddle River, NJ: Merrill Prentice Hall.
Paivio, A. (1971). Mental representations: A dual-coding approach. New York, NY: Oxford University Press.
Paivio, A. (1986). Mental Representations. New York, NY: Oxford University Press.
Stantrock, J. W. (2003). Psychology (7th ed.). Boston: McGraw-Hill.
Schunk, D. H. (2004). Learning theories: An educational perspective (4th ed.). Upper Saddle River, NJ: Pearson.
Sekuler, B. & Kahana, M. (n.d.). Understanding visual memory. The Volen Center at Brandeis University Web site: http://people.brandeis.edu/~sekuler/MemoryDemo/
Terry, W. S. (2000). Learning and memory: Basic principles, processes and procedures. Boston, MA: Allyn and Bacon.
Wagner, A. D., Schacter, D. L., Rotte, M., Koutstaal, W., Maril, A., Dale, A. M., Rosen, B. R. & Buckner, R. L. (1998). Building memories: Remembering and forgetting of verbal experiences as predicted by brain activity. Science, 281, 1188-1191.