Cascella Marco, Al Khalili Yasir
Istituto Nazionale Tumori - IRCCS - Fondazione Pascale, Via Mariano Semmola 80100, Napoli. Italy
Virginia Commonwealth University
Short-term memory is also called short-term storage, primary memory, or active memory. The term indicates different systems of memory involved in retaining pieces of information, or memory chunks, for a relatively short time, typically up to 30 seconds. In contrast, long-term memory may hold indefinite information. However, the difference is not just the time variable but also their overall functionality. Nevertheless, the 2 systems are closely related. Practically, short-term memory functions as a temporary scratchpad for recalling a limited amount of data, typically around 7 ± 2 items, in the verbal domain, based on George Miller's concept. This information originates from the sensory register and can be processed through attention and recognition. In contrast, information collected in long-term memory storage consists of memories for performing actions or skills, such as procedural memories (knowing how), and memories of facts, rules, concepts, and events, such as declarative memories (knowing that). Declarative memory includes semantic and episodic memory. The former concerns broad knowledge of facts, rules, concepts, and propositions (general knowledge), whereas the latter concerns personal and experienced events and the contexts in which they occurred (personal recollection). Although short-term memory is closely related to the concept of working memory, both are distinct entities. Short-term memory is a set of storage systems, whereas working memory indicates the cognitive operations and executive functions associated with the organization and manipulation of stored information. Nevertheless, the terms short-term memory and working memory are often used interchangeably. Short-term memory must also be distinguished from sensory memory, such as the acoustical echoic and iconic visual memories, which are shorter in duration, typically fractions of a second, and reflect the stimulus's original sensation or perception. In other words, sensory memory is specific to the stimulus modality of presentation. This raw sensory information undergoes processing and, upon processing to short-term, is expressed in a format different from that perceived initially. The famous Atkinson and Shiffrin (or multi-store) model, proposed in the late 1960s, explains the functional correlations between different types of memory. Many studies demonstrated the anatomical and functional distinctions between memory processes, neural correlates, and the functioning of short-term and long-term memory subsystems. In light of these findings, several memory models have been postulated. Although some authors suggested a single memory system encompassing both short- and long-term storage after 50 years, the Atkinson and Shiffrin model remains a valid approach for explaining the memory dynamics. In light of recent research, however, the model has several problems, mostly concerning the characteristics of short-term memory, the relationship between short-term and working memory, and the transition from short- to long-term memory. Short-term memory is a storage system that includes several subsystems with limited capacity. Rather than a limitation, this restriction is an evolutionary survival advantage, as paying attention to limited but essential information, excluding confounding factors, is important. A classic example in biology is prey, which must focus on the hostile environment to recognize a possible attack by the predator. Given the functional peculiarities of short-term memory, which involves the collection of sensory information, the subsystems are closely related to the modalities of sensory memory. Consequently, several sensory-associated subsystems have been postulated, including the visuospatial, phonological (auditory-verbal), tactile, and olfactory domains. These subsystems involve patterns and functional interconnections with the corresponding cortical and subcortical areas and centers. In 1974, Baddeley and Hitch developed an alternative model called working memory. This model does not exclude the modal model but rather enriches the contents. The short-term store can be used to characterize the functioning of the working memory. Working memory refers to the entire theoretical framework of the structures and processes used for storing and temporarily manipulating information, of which short-term memory is only a component. In other words, short-term memory is a functional storage element, whereas working memory is a set of processes involving storage phases. Working memory is constantly used when we have to understand new information, solve a problem, or make an argument, and it is the cognitive strategy for achieving short-term goals. The importance of this type of operating system of memory is demonstrated by the evidence showing that working memory deficits are associated with several developmental disorders of learning, including attention-deficit hyperactivity disorder, dyslexia, and specific language impairment. A recent study on working and short-term memory in children with attention-deficit hyperactivity disorder suggested that these children had an inadequate hemodynamic response in a region of the brain that underlies phonological working memory. Functional near-infrared spectroscopy is a cost-effective, noninvasive neuroimaging technique that localizes and quantifies neural activation patterns related to executive functions. Another study suggested that working memory impairment and attention lapsing are general features of psychotic disorders. Findings regarding capacity estimates from the Change Localization and Detection tasks related to functional capacity and outcome also implied that these methods may be useful in a clinical context. Short-term and long-term memory can be distinguished based on storage capacity and duration. Short-term memoryhas limitations in the amount and duration of information maintained. In contrast, long-term memory features a seemingly unlimited capacity that can last years. The functional distinctions between memory storage systems and the exact mechanisms for how memories transfer from short-term to long-term memory remain controversial. Do short- and long-term memory represent 1 or more systems with specific subsystems? Although short-term memory probably represents a substructure of long-term memory, a form of long-term–activated storage, rather than looking for a physical division, verifying the mechanisms involved in transitioning from short-term to long-term memory seems appropriate. Although the classic multi-modal model proposed that the storage of short-term memories occurs automatically without manipulation, the matter seems more involved. The phenomenon concerns quantitative (number of memories) and qualitative (quality of memory) features. Regarding quantitative data, although the number of Miller of 7 ± 2 items identifies the number of elements included among individual slots, grouping memory bits into larger chunks (chunking) can allow for greater storage capacity. The qualitative issue, or memory modulation within processing, is fascinating. Short-term memory elements undergo processing, providing a sort of editing involving each element's fragmentation (chunking) and re-elaboration. This phase of memory processing is called encoding and can condition subsequent processing, including storage and retrievalThe encoding process encompasses automatic (without conscious awareness) and effortful processing (through attention, practice, and thought) and allows us to retrieve information to make decisions and answer questions. Three pathways are followed during the encoding step—visual (information represented as a picture), acoustic (information represented as a sound), and semantic encoding (the meaning of the information). These processes interconnect, breaking down information into various components. During recovery, the pathway that has produced the coding facilitates the recovery of the other components through a chain reaction. A particular perfume, for instance, makes us recall a specific episode or image. The encoding process affects the recovery, but the recovery undergoes potential changes that can alter the initial content. In neurofunctional terms, the difference is the occurrence in long-term memory of a series of events that must definitively fix the engram(s). This effect occurs through the establishment of neural networks and is expressed as a neurofunctional phenomenon, including long-term potentiation, which is an increase in the strength of the neural transmission deriving from the strengthening of synaptic connections. This process requires gene expression and the synthesis of new proteins and is related to long-lasting structural alterations in the synapses (synaptic consolidation) of the brain areas involved, such as the hippocampus in declarative memories. Hippocampal neurogenesis regulates the maintenance of long-term potentiation. However, the hippocampal network, including the parahippocampal gyrus, hippocampus, and neocortical areas, is not the storage site for memories but plays a crucial role in forming new memories and their subsequent reactivation. The hippocampus appears to have a limited capacity but acquires information quickly and automatically. Over time, the initially available information is permanent in other brain structures, particularly the cortex), independently from the activity of the hippocampus itself. The crucial mechanism of this transfer involves the reactivation (replay) of neural activity configurations. In other words, the connected hippocampus and the medial temporal structures are crucial for holding an event as a whole as they distribute memory traces in an organized manner. The operating system can store, organize, process, and recover hardware files through different software. This hippocampal-guided reactivation (retrieval) leads to the creation of direct connections between the cortical traces and then to the formation of an integrated representation in the neocortex, including the visual association cortex for visual memory, the temporal cortex for auditory memory, and the left lateral temporal cortex for knowledge of word meanings. Moreover, the hippocampus has other specific tasks, such as spatial memory organization. Recent reviews summarized the progress of hippocampal circuits and functions based on sharp-wave ripples. These ripples are crucial for consolidating spatial, episodic, and social memories in different hippocampal-cortical pathways. Dysregulation of sharp-wave ripples contributes to cognitive impairments in neurodegenerative and neurodevelopmental diseases. Other brain areas are involved in memory processes; for example, learning motor skills has links to the activation of the cerebellar regions and brainstem nuclei. Learning perceptive activities, including improvements in processing perceptive stimuli essential in everyday life activities such as understanding spoken and written language, involves basal ganglia and sensory and associative cortices. In contrast, learning cognitive skills related to problem-solving involves the medial temporal lobes.
短期记忆也被称为短期存储、初级记忆或主动记忆。该术语表示参与在相对较短时间内(通常最长30秒)保留信息片段或记忆组块的不同记忆系统。相比之下,长期记忆可以保存无限量的信息。然而,区别不仅在于时间变量,还在于它们的整体功能。尽管如此,这两个系统密切相关。实际上,根据乔治·米勒的概念,短期记忆在言语领域起着临时便签簿的作用,用于回忆有限数量的数据,通常约为7±2个项目。这些信息源自感觉登记,可以通过注意力和识别进行处理。相比之下,长期记忆存储中收集的信息包括执行动作或技能的记忆,如程序性记忆(知道如何做),以及事实、规则、概念和事件的记忆,如陈述性记忆(知道是什么)。陈述性记忆包括语义记忆和情景记忆。前者涉及对事实、规则、概念和命题的广泛知识(常识),而后者涉及个人经历的事件以及这些事件发生的背景(个人回忆)。虽然短期记忆与工作记忆的概念密切相关,但两者是不同的实体。短期记忆是一组存储系统,而工作记忆则表示与存储信息的组织和操作相关的认知操作和执行功能。然而,短期记忆和工作记忆这两个术语经常互换使用。短期记忆也必须与感觉记忆区分开来,如听觉回声记忆和视觉图像记忆,它们的持续时间更短,通常只有几分之一秒,并且反映刺激的原始感觉或感知。换句话说,感觉记忆特定于呈现的刺激模态。这种原始的感觉信息经过处理,在处理为短期记忆时,以与最初感知不同的格式表达。20世纪60年代末提出的著名的阿特金森和谢夫林(或多存储)模型解释了不同类型记忆之间的功能相关性。许多研究证明了记忆过程、神经相关性以及短期和长期记忆子系统功能之间的解剖学和功能差异。鉴于这些发现,人们提出了几种记忆模型。尽管一些作者在50年后提出了一个包含短期和长期存储的单一记忆系统,但阿特金森和谢夫林模型仍然是解释记忆动态的有效方法。然而,根据最近的研究,该模型存在几个问题,主要涉及短期记忆的特征、短期和工作记忆之间的关系以及从短期到长期记忆的转变。
短期记忆是一个存储系统,包括几个容量有限的子系统。这种限制并非缺陷,而是一种进化生存优势,因为关注有限但重要的信息,排除混杂因素很重要。生物学中的一个经典例子是猎物,它必须专注于敌对环境以识别捕食者可能的攻击。鉴于短期记忆的功能特性,即涉及感觉信息的收集,这些子系统与感觉记忆的模态密切相关。因此,人们提出了几个与感觉相关的子系统,包括视觉空间、语音(听觉 - 言语)、触觉和嗅觉领域。这些子系统涉及与相应皮质和皮质下区域及中枢的模式和功能连接。
1974年,巴德利和希契开发了一种名为工作记忆的替代模型。该模型并不排除模态模型,而是丰富了其内容。短期存储可用于描述工作记忆的功能。工作记忆是指用于存储和临时操纵信息的结构和过程的整个理论框架,短期记忆只是其中一个组成部分。换句话说,短期记忆是一个功能性存储元素,而工作记忆是一组涉及存储阶段的过程。当我们必须理解新信息、解决问题或进行论证时,会不断使用工作记忆,它是实现短期目标的认知策略。有证据表明工作记忆缺陷与几种学习发育障碍有关,包括注意力缺陷多动障碍、阅读障碍和特定语言障碍,这证明了这种记忆操作系统的重要性。最近一项关于注意力缺陷多动障碍儿童的工作记忆和短期记忆的研究表明,这些儿童在语音工作记忆所依赖的大脑区域存在血液动力学反应不足的情况。功能近红外光谱是一种经济高效的非侵入性神经成像技术,可定位和量化与执行功能相关的神经激活模式。另一项研究表明,工作记忆损害和注意力不集中是精神疾病的普遍特征。与功能能力和结果相关的变化定位和检测任务的容量估计结果也表明,这些方法在临床环境中可能有用。
短期记忆和长期记忆可以根据存储容量和持续时间来区分。短期记忆在信息保持的数量和持续时间上存在限制。相比之下,长期记忆具有看似无限的容量,可以持续数年。记忆存储系统之间的功能差异以及记忆从短期向长期转移的确切机制仍然存在争议。短期记忆和长期记忆是代表一个还是多个具有特定子系统的系统?虽然短期记忆可能代表长期记忆的一个子结构,一种长期激活的存储形式,但与其寻找物理划分,验证从短期到长期记忆转变所涉及的机制似乎更合适。虽然经典的多模态模型提出短期记忆的存储是自动发生的,无需操纵,但情况似乎更为复杂。这个现象涉及定量(记忆数量)和定性(记忆质量)特征。关于定量数据,虽然米勒提出的7±2个项目确定了各个槽位中包含的元素数量,但将记忆位组合成更大的组块(组块化)可以允许更大的存储容量。定性问题,即处理过程中的记忆调制,很有趣。短期记忆元素经过处理,提供一种编辑,涉及每个元素的碎片化(组块化)和重新加工。记忆处理的这个阶段称为编码,它可以影响后续处理,包括存储和检索。编码过程包括自动(无意识)处理和努力处理(通过注意力、练习和思考),并使我们能够检索信息以做出决策和回答问题。编码步骤遵循三条途径——视觉(信息表示为图片)、听觉(信息表示为声音)和语义编码(信息的含义)。这些过程相互关联,将信息分解为各种组件。在恢复过程中,产生编码的途径通过连锁反应促进其他组件的恢复。例如,一种特定的香水会让我们回忆起一个特定的事件或图像。编码过程会影响恢复,但恢复过程可能会发生潜在变化,从而改变初始内容。从神经功能角度来看,区别在于长期记忆中必须最终固定记忆痕迹的一系列事件的发生。这种效应通过神经网络的建立而发生,并表现为一种神经功能现象,包括长期增强,即由于突触连接的加强而导致神经传递强度的增加。这个过程需要基因表达和新蛋白质的合成,并且与大脑区域(如陈述性记忆中的海马体)突触的持久结构改变(突触巩固)有关。海马体神经发生调节长期增强的维持。然而,包括海马旁回、海马体和新皮质区域在内的海马体网络不是记忆的存储部位,而是在形成新记忆及其随后的重新激活中起关键作用。海马体似乎容量有限,但能快速自动获取信息。随着时间的推移,最初可用的信息在其他大脑结构(特别是皮质)中永久存在,独立于海马体自身的活动。这种转移的关键机制涉及神经活动配置的重新激活(回放)。换句话说,相连的海马体和内侧颞叶结构对于将一个事件作为一个整体进行保存至关重要,因为它们以有组织的方式分布记忆痕迹。操作系统可以通过不同的软件存储、组织、处理和恢复硬件文件。这种由海马体引导的重新激活(检索)导致皮质痕迹之间建立直接连接,然后在新皮质中形成整合表示,包括视觉记忆的视觉联合皮质、听觉记忆的颞叶皮质和单词意义知识的左侧颞叶皮质。此外,海马体还有其他特定任务,如空间记忆组织。最近的综述总结了基于尖波涟漪的海马体回路和功能的进展。这些涟漪对于在不同的海马体 - 皮质途径中巩固空间、情景和社会记忆至关重要。尖波涟漪的失调会导致神经退行性疾病和神经发育疾病中的认知障碍。其他脑区也参与记忆过程;例如,学习运动技能与小脑区域和脑干核的激活有关。学习感知活动,包括改善日常生活活动(如理解口语和书面语言)中至关重要的感知刺激处理,涉及基底神经节以及感觉和联合皮质。相比之下,学习与解决问题相关的认知技能涉及内侧颞叶。
2025-1
Psychopharmacol Bull. 2024-7-8
Arch Ital Urol Androl. 2025-6-30
2025-1
Autism Adulthood. 2024-12-2
Health Technol Assess. 2001