Abstract

Working memory and consciousness are crucial for everyday functioning and are intimately related. Over a century of research has explored this relationship, from the early works of Fechner and Wundt, through to predominant current-day theories of Baddley’s Working Memory (WM) and Baars’s Global Workspace Theory (GWT). WM, originally proposed to consist of a visuospatial sketchpad, phonological loop, and central executive, is more fractionated than originally envisaged, with discrete areas for different sensory representations integrated into coherent, perceived ‘wholes’ via a multi-representational transient episodic buffer. GWT builds on this, articulating an explicit role for consciousness. GWT proposes that the ‘spotlight of attention’ selects stimuli for conscious awareness which are then broadcast widely across the whole brain, and made available for processing by largely unconscious functional processors. GWT is confirmed by a strong empirical and neuroscientific base and has increased understanding of brain-based illnesses and psychopathology. This leads to two important observations: conscious-awareness and psychopathology may be influenced by WM and whole-brain health; and most of ‘what we do’ may be determined by unconscious processes. Attention itself has also been found to be fractionated, leading to multiple possibilities for fruitful future research.

Keywords: working memory, global workspace theory, attention, consciousness

Working Memory and Consciousness: A Review of Theory and Research Directions

Working memory plays a vital role in numerous aspects of daily life, including  storing and recalling information, interpreting and attending to sensations, learning and concentrating (Goldstein, 2015), and regulating emotions (Rutherford, Booth, Crowley, & Mayes, 2015). Similarly, consciousness appears vital to human survival, facilitating knowledge of and interaction between self and the environment (Marchetti, 2018). Working memory and consciousness seem intimately related, and possibly equal (Baddeley, 2007). Current research attempts to more fully explicate this relationship, which affects many domains including perception, attention, cognition, learning, brain-injury recovery, and others (Velichkovsky, 2017; Overgaard, 2017). This essay explores the theoretical foundations of the relationship from its roots in the early works of Fechner, Ebbinghaus and Wundt, through to two prominent current-day theories: Working Memory (WM) (Baddeley, 1986; 200; 2012), and Global Workspace Theory (GWT) (Baars, 2003; 2005). It finds strong empirical support for their integration, even though WM is fractionated beyond the original visuo-spatial sketchpad and phonological loop envisaged by Baddeley, with discrete areas having been identified for other sensory processing including kinaesthetic, olfactory, emotional, and haptic/tactile (Smyth & Pendleton, 1990; Dade et-al.; 2001; Mikels, et-al, 2008; Katus & Eimer 2018; Schmidt & Blankenburg; 2018). Consciousness is articulated as particular elements of WM —the Central Executive (CE) and Episodic Buffer (EB)—illuminated by the ‘spotlight of attention’; however, there may be more than one ‘spotlight’, with separate attentional mechanisms having been identified for different sensory processes. Based on recent research incorporating GWT, this paper notes that whole-brain and WM health potentially influence both conscious awareness and psychopathology, and identifies potential future areas for research.

Gustav Fechner laid the foundations of modern cognitive psychology when, in 1860, he published Elemente der Psychophsik  (Link, 2005). Fechner regarded body and soul as dual manifestations of the same being, and asked “what is the relation between the intensity of a psychical [sic] action—estimated by consciousness—and the strength of the underlying physical action, as measured by the work done?” (Ward, 1876, p. 452). Fechner viewed the sensory system as the mind’s “connective tunnel” to the external world, and proposed that sensations derive from transforming the energy of physical stimuli into neural energy. Fechner’s Law states subjective sensation is proportional to the logarithm of the stimulus intensity and from this, “the future of experimental psychology was set” (Link, 2005).

Donders, Ebbinghaus and Wundt followed on Fechner’s heels. In 1868, by measuring physical reaction time, Dutch physiologist Franciscus Donders inferred how long it takes to make decisions, and is credited as the first person to measure inferred mental processes through measuring physical behaviour (Goldstein, 2015). Meanwhile, Hermann Ebbinghaus was exploring principles of memory, learning, and recall, and his work on forgetting curves was a precursor to modern research on working memory (Goldstein, 2015). Wilhelm Wundt, best known for structuralism and analytic introspection, was also deeply interested in the study of the mind and its relationship to consciousness (Cowan & Rachev, 2018). Indeed, his ‘An Introduction to Psychology’ describes psychology as an empirical science that must “investigate the facts of consciousness, its combinations and relations, so that it may ultimately discover the laws which govern these relations and combinations” (Wundt, 1912, p. 15). Wundt’s model of memory and consciousness is below.

With the rise of behaviourism, the study of the mind diminished through the first half of the 20^th century, until the cognitive revolution in the 1950s (Goldstein, 2015). Since then, significant progress has been made towards understanding memory and consciousness.

The currently dominant model of WM is based on the work of Baddeley (1986; 1992; 2000; 2010; 2012), with over 50,000[1] citations; 7,500 since 2017. Adopting a view that theories should be like maps to guide further exploration, rather than complete explications, Baddeley spent the 1960s working from the ‘maps’ of others—primarily Broadbent, Atkinson, Shiffrin, and Toulman—and finally, in 1974, published Baddeley and Hitch’s first working memory model (Baddeley, 2012), shown below:

The model contains two Short Term Memory (STM) areas for sensory data—the visuo-spatial sketch pad and the phonological loop—linked with a ‘Central Executive’ whose main function is the attentional control of action. Baddeley and other researchers demonstrated through multiple experiments—the phonological similarity effect, word length effect, articulatory suppression and others—that STM areas have limited capacity, and data can be held for only a short time before being either displaced or overwritten. Baddley initially treated STM and Long-Term Memory (LTM) as separate, but later contemplated direct links between WM and LTM, with bidirectional flow of information (Baddeley, 2012). The model was updated in 1986, and again in 2000, to show these linkages and to distinguish between WM as ‘fluid systems’, needing only temporary activation, and LTM as ‘crystallized’ knowledge and skills. The role of the CE was further explicated as focusing attention, dividing attention, switching between tasks, and interfacing with LTM. To accommodate evidence that the CE is not solely an attentional system, but also has discrete storage capacity, another element called the Episodic Buffer (EB) was added to the model (Baddeley, 2012).

Subsequent research has demonstrated fractionation of the visuo-spatial sketchpad into separate areas, including visual (Logie, 1986), spatial (Baddeley & Lieberman, 1980), kinaesthetic/movement-based (Smyth & Pendelton, 1990), olfactory (Dade, et-al, 2001); emotional (Mikels et-al, 2008); and haptic/tactile (Katus & Eimer, 2018; Schmidt & Blankenburg, 2018). Each area is specific to its respective function; however, the EB holds multidimensional representations, buffering storage for all other areas and linking together WM, perception, and LTM. It is this episodic buffer that bridges memory and consciousness (Baddeley, 2007).

Although a theory of consciousness is not explicit in Baddeley’s model, he writes extensively on the topic (Baddeley, 2007; 1992), and it is implicit in the WM model (Velichkovsky, 2017). In summary, WM holds representations that, by the focusing of attention, are consciously experienced and accessed via the EB and CE, which bind together stimuli from various other—non-conscious—components into integrated, perceived wholes. Key points articulated by Baddeley include that: cognitive processing is possible without consciousness (e.g. blindsight, prosopagnosia, and hemineglect); some cognitive operations do require consciousness; attention is necessary for consciousness; and attention, directed by the CE and facilitated by the EB, mobilises and amplifies neural processes involved with information processing and action (Baddeley, 2010; 2007; 2012). In this sense, Baddeley’s WM is a system for storing and manipulating information which incidentally, but importantly, contains components that underpin consciousness.

Baars picks up on this incidental connection, building on the WM model to incorporate consciousness explicity, articulated in Global Workspace Theory (GWT) (Baars & Franklin, 2003; Baars, 2005). Below, the two models are shown for comparison (from Baars, 2003):

At first sight the two models appear to broadly agree, but GWT contains important modifications. Firstly, it encapsulates the conscious elements of WM—the Central Executive and Episodic Buffer from Baddeley’s model—in the Global Workspace of Baars’s model. Secondly, it proposes that stimuli are selected by consciousness via the ‘spotlight of attention’ and distributed, or ‘broadcast’ widely across (unconscious) specialised areas of the brain. These areas then ‘pick up’ the broadcast data that is ‘relevant’ to their specialised processing capabilities.

The notion of a ‘spotlight of attention’ is central to GWT. From myriad available stimuli, the focusing of attention selects the few to be made universally available—broadcast—across the brain, to be then selected by unconscious cognitive processors. Baars uses the metaphor of a theatre spotlight which shows only a fraction of the stage at a time;  other areas are dimly lit, and the audience is only vaguely aware of their contents until the spotlight of attention brings them into conscious awareness. Reductionist views, suggesting consciousness is equivalent to attention, include those of Prinz (2010) and Mole (2011); others such as Breitmeyer (2014) demur; and keen debate continues.

Meanwhile, cognitive psychology has turned its eye to the pluralistic nature of attention (Taylor, 2015), with evidence suggesting attention itself may be fractionated. Daryl Fougnie (2008) found distinct psychological systems for visuospatial attention and central attention. More recently, independent attention mechanisms have been found to control activation of visual and tactile representations in WM (Katus & Eimer, 2018). Additionally, all attention is not equal: variable weightings are apparently allocated by attention, with higher-value items being held in WM for longer (Atkinson, et-al., 2018); however these higher-value items have greater fragility and vulnerability to interference (Allen & Ueno, 2018).

GWT emerged from and is supported by a strong empirical base in cognitive science and neuroscience (Baars, 2005). Neuroscientific evidence indicates that conscious awareness involves widely distributed activation across multiple brain regions. For example, Godwin, Barry, and Marois (2015) observed “whole-brain awareness” when subjects were asked to observe an image flashed briefly on a screen. Research into epilepsy also supports GWT (Bartolomei & Naccache, 2011) with several lines of evidence that seizures result from disturbances within the global workspace, offering new perspectives on treatment (Bartolomei, McGonigal, & Naccache, 2014). Understanding of psychopathology has similarly advanced through the lens of WM and GWT; for example in anorexia nervosa (Boehm, et-al., 2018), emotional regulation (Smith & Lane, 2016), substance abuse (Brooks, 2016), and schizophrenia (Berkovitch, et-al, 2018).

From a cognitive psychology viewpoint, two important observations arise: firstly, that conscious awareness—how we experience the world—and possibly the resolution of psychological disorders, rest, to some extent on whole-brain and WM health; secondly, that conscious awareness represents only a fraction of the cognitive cycle, with most pertinent activity occurring unconsciously. That is, most of what leads to “what we do” is unconscious.

Discussion

Through a century of empirical and neuroscientific research, the current overall understanding of memory and consciousness is remarkably in accord with Wundt’s 1912 model (Figure 1), albeit with considerably greater explication and detail (Cowan & Rachev, 2018). GWT incorporates Wundt’s ‘unknowable unconscious’ as discrete, functional, unconscious processors widely distributed across the brain yet vastly interconnected, as supported by robust neuroscientific evidence. Wundt’s ‘field of consciousness’ and ‘focus of attention’ closely resemble GWT’s ‘spotlight of attention’ illumiating the combined workings of the transient EB and CE.

GWT affords an explanation of consciousness with WM at its base. WM is evidently fractionated beyond the original model of visuospatial workpad and phonological loop, with additional domain-specific areas for tactile, haptic, olfactory, emotional, visual, and spatial processing—collectively, the global workspace. Consciousness arises from attending to selected stimuli, whose representations are then broadcast throughout the global workspace, to trigger various unconscious functional processors. Strong evidence supports GWT, and has led to new possibilities for treating brain-based illnesses.

Emerging research shows that attention is also fractionated, and this is likely to cultivate fruitful fields of future research. Possibilities include deepening our understanding of individual attentional mechanisms; the how and why of selecting stimuli to be broadcast; and what triggers discrete unconscious processes to respond to stimuli from the global broadcast. Building on the work of Atkinson et-al. (2018) and Allen & Ueno (2018), research should proceed on the prioritisation processes of the separate attentional mechanisms within WM. This might be interpretable in terms of allocating resources based on competing wants and needs, learned or hard-wired: that is, the unconscious processor recipients of global workspace broadcasts might well stand for contingent states of disequilibria, whose priorities for resolution vary according to the context of stimuli being broadcast.

We may acknowledge Wundt’s urging in such light, to “investigate the facts of consciousness, its combinations and relations, to ultimately discover the laws which govern these relations and combinations”, to benefit human wellbeing.

References

Allen, R., & Ueno, T. (2018). Multiple high-reward items can be prioritized in working memory but with greater vulnerability to interference. Attention, Perception, and Psychophysics, 80(7), 1731-1743. doi:10.3758/s13414-018-1543-6

Atkinson, A., Berry, E., Waterman, A., Baddeley, A., Hitch, G., & Allen, R. (2018, Jul). Are there multiple ways to direct attention in working memory? Annals of the New York Academy of Sciences, 14241(1), 115-126. doi:10.1111/nyas.13634

Baars, B. (2005). Global workspace theory of consciousness: Toward a cognitive neuroscience of human experience. Progress in Brain Research, 150, 45-53. doi:10.1016/S0079-6123(05)50004-9

Baars, B., & Franklin, S. (2003, Apr). How conscious experience and working memory interact. Trends in Cognitive Sciences, 7(4), 166-172. doi:10.1016/S1364-6613(03)00056-1

Baddeley, A. (1986). Working Memory. Oxford: Oxford University Press.

Baddeley, A. (1992). Consciousness and working memory. Consciousness and Cognition, 1(1), 3-6. doi:10.1016/1053-8100(92)90037-B

Baddeley, A. (2000). The episodic buffer: A new component of working memory? Trends in Cognitive Sciences, 4(11), 417-423.

Baddeley, A. (2007). Consciousness. In Working memory, thought, and action. Oxford: Oxford University Press. Retrieved from https://ebookcentral.proquest.com/lib/cdu/detail.action?docID=1336454

Baddeley, A. (2010). Primer: Working Memory. Current Biology, 20(4), 136-141. doi:10.1016/j.cub.2009.12.014

Baddeley, A. (2012). Working Memory: Theories, Models, and Controversies. Annual Review of Psychology, 63, 1-29. doi:10.1146/annurev-psych-120710-100422

Baddeley, A., & Lieberman, K. (1980). Spatial working memory. In R. Nickerson (Ed.), Attention and Performance (Vol. 3, pp. 521-39). New Jeresy: Erlbaum.

Bartolomei, F., & Naccache, L. (2011). The Global Workspace (GW) Theory of Consciousness and Epilepsy. Behavioural Neurology, 24(1), 67-74. doi:10.3233/BEN-2011-0313

Bartolomei, F., McGonigal, A., & Naccache, L. (2014, Jan). Alteration of consciousness in focal epilepsy: The global workspace alteration theory. Epilepsy and Behaviour, 30, 17-13. doi:10.1016/j.yebeh.2013.09.012

Berkovitch, L., Del Cul, A., Maheu, M., & Dehaene, S. (2018). Impaired conscious access and abnormal attentional amplification in schizophrenia. NeuroImage, 18, 835-848. doi:10.1016/j.nicl.2018.03.010

Boehm, I., King, J., Bernardoni, F., Geisler, D., Seidel, M., Ritschel, F., . . . Ehrlich, S. (2018). Subliminal and supraliminal processing of reward-related stimuli in anorexia nervosa. Psychological Medicine, 48(5), 790-800. doi:10.1017/S0033291717002161

Breitmeyer, B. (2014). Consciousness and attention: Partners but not equal. In The Visual (Un)Conscious and Its (Dis)Contents: A microtemporal approach. Oxford Scholarship Online. doi:10.1093/acprof:oso/9780198712237.001.0001

Brooks, S. J. (2016). A debate on working memory and cognitive control: can we learn about the treatment of substance use disorders from the neural correlates of anorexia nervosa? BMC Psychiatry, 16(10). doi:10.1186/s12888-016-0714-z

Cowan, N., & Rachev, N. R. (2018, June). Merging with the path not taken: Wilhelm Wundt’s work as a precursor to the embedded-processes approach to memory, attention, and consciousness. Consciousness and Cognition, 63, 228-238. doi:10.1016/j.concog.2018.06.001

Dade, L., Zatorre, E., Evans, A., & Jones-Gottman, M. (2001). Working memory in another dimension: Functional imaging of human olfactory working memory. Neuroimage, 14, 650-660. doi:10.1006/nimg.2001.0868

De Brigard, F., & Prinz, J. (2010). Attention and consciousness. Interdisciplinary Reviews-Cognitive Science, 51-59. doi:10.1002/wcs.27

Fougnie, D. (2008). The Relationship Between Attention and Working Memory. In N. Johansen (Ed.), New Research on short-term memory. New York: Nova Biomedical Books.

Godwin, D., Barry, R., & Marois, R. (2015). Breakdown of the brain’s functional network modularity with awareness. Proceedings of the National Academy of Sciences of the United States of America, 112(12), 3799. doi:10.1073/pnas.1414466112

Goldstein, E. (2015). Cognitive Psychology: Connecting Mind, Research and Everyday Experience (4th ed.). Cengage Learning.

Katus, T., & Eimer, M. (2018, Jan). Independent Attention Mechanisms Control the Activation of Tactile and Visual Working Memory Representations. Journal of Cognitive Neuroscience, 30(5), 644-655. doi:10.1162/jocn_a_01239

Link, S. (2005). Introductory Article. Fechner, Gustav Theodor, 1-4. Retrieved from http://psy2.ucsd.edu/~link/Link%20fechner%20ECS%202006.pdf

Logie, R. (1986). Visuo-spatial processing in working memory. Quarterly Journal of Experimental Psychology, 38A, 229-247. doi:doi.org/10.1080/14640748608401596

Marchetti, G. (2018). Consciousness: a unique way of processing information. Cognitive Processing, 19, 435–464. doi:10.1007/s10339-018-0855-8

Mikels, J., Reuter-Lorenz, P., Beyer, J., & Fredrickson, B. (2008). Emotion and working memory: evidence for domain-specific processes for affective maintenance. Emotion, 256-266. doi:10.1037/1528-3542.8.2.256

Mole, C. (2011). Attention is Cognitive Unison. New York: Oxford University Press.

Overgaard, M. (2017). The Status and Future of Consciousness Research. Frontiers in Psychology, 8(1719). doi:10.3389/fpsyg.2017.01719

Rutherford, H. J., Booth, C. R., Crowley, M. J., & Mayes, L. C. (2015, Sep). Investigating the relationship between working memory and emotion regulation in mothers. Journal of Cognitive Psychology, 28(1), 52-59. doi:10.1080/20445911.2015.1075542

Schmidt, T., & Blankenburg, F. (2018, May). Brain regions that retain the spatial layout of tactile stimuli during working memoray – A ‘tactospatial sketchpad’? NeuroImage, 178, 531-539. doi:10.1016/j.neuroimage.2018.05.076

Smith, R., & Lane, R. D. (2016). Unconscious emotion: A cognitive neuroscientific perspective. Neuroscience and Biobehavioral Reviews, 69, 216-238. doi:10.1016/j.neubiorev.2016.08.013

Smyth, M., & Pendleton, L. (1990). Space and movement in working memory. The Quarterly Journal of Experimental Psychology, 42, 291-304. doi:10.1080/14640749008401223

Taylor, J. (2015). Against Unifying Accounts of Attention. Erkenntnis, 80(1), 39-56. doi:10.1007/s10670-014-9611-3

Velichkovsky, B. B. (2017, Oct). Consciousness and working memory: Current trends and research. Consciousness and Cognition, 55, 35-45. doi:10.1016/j.concog.2017.07.005

Ward, J. (1876). An Attempt to Interpret Fechner’s Law. Mind, 1(4), 452-466. Retrieved from www.jstor.org/stable/2246409

Wundt, W. (1912). An Introduction to Psychology. (R. Pintner, Trans.) New York: The Macmillan Company. Retrieved Sep 25, 2018, from https://ia600304.us.archive.org/18/items/introductiontops1912wund/introductiontops1912wund_bw.pdf

---

[1] Google Scholar search, 25/09/2018