What is the best way to memorise words?

Human beings never stop learning new words. Often this happens without any special effort; therefore, we tend to take it for granted. Nonetheless, the mechanisms used to learn vocabulary remain one of the top mysteries of cognitive neuroscience. Researchers from the University embraced the challenge of solving this mystery.

‘We focus on the ways to learn new words. There are two main options that are described in the literature, and in practice everyone is familiar with them. Explicit learning is the way we study vocabulary at school and in foreign language classes. We are shown an object and the teacher says: look, this is called “so and so.” Another way is implicit learning, when we acquire new knowledge based on the context,’ explained Olga Shcherbakova, Associate Professor at St Petersburg University (Department of General Psychology).

Material from the magazine ‘St Petersburg Universityъ

According to Olga Shcherbakova, presently there are conflicting data regarding the effective methods of learning and the neuronal mechanisms that support learning in each case. There is evidence that the explicit mode of learning is based on slower subcortical processes, while the implicit one is supported by neocortical mechanisms. However, other results were obtained that contradict these data. ‘The fact is that neuronal mechanisms are very difficult to describe. The data available today were gathered mainly through the functional Magnetic Resonance Imaging at high spatial resolution. It enables us to register the neuronal shifts that occur at a low speed. However, we cannot track the so-called immediate neuroplastic changes, which may be crucial to the process of gaining understanding and meaning of the new words,’ Olga Shcherbakova pointed out. ‘Also, there have been serious concerns about the experimental protocols of the previous research.’

Pseudowords for experiment

To bring the study of verbal learning to another level, researchers from the University undertook to develop an experimental paradigm that would be carefully balanced with respect to the cognitive load, as well as the auditory and visual properties of the stimuli presented to the subject.

We started by selecting 20 Russian words. These were quite frequently used nouns denoting objects and animals. All these words are triphones and trigrams; that is, they consist of three sounds and three letters – for instance, ‘фен’ /fen/ (a hairdryer) or ‘мул’ /mul/ (a mule). Despite the wealth of the Russian language, this task proved to be very difficult.

Associate Professor Olga Shcherbakova, St Petersburg University (Department of General Psychology)

Afterwards the researchers took the beginning and the end from each real word and made up several hundred pseudowords by rearranging the pieces. Of this number, only those pseudowords were selected that do not sound similar to the words which already exist in the Russian language (otherwise they would be involuntarily perceived as meaningful and familiar). The resulting list included pseudowords that do not exist in Russian, yet follow its rules, for example, ‘фел’ /fel/ or ‘мун’ /mun/. ‘Admittedly, such work has been rarely done before. Therefore, our results have an independent value,’ emphasised the researcher.

Each pseudoword was tested in five different experimental conditions – training sessions of a sort. The word could mean an object or an animal in explicit and implicit learning modes, or it could serve as a meaningless filler word with no semantic value assigned to it. ‘The research subjects listened to these words passively while watching a cartoon, without performing any task. We wanted to find out whether there is a difference between memorising words with a meaning and words that are devoid of meaning,’ clarified Olga Shcherbakova. After the training sessions, the participants performed three tasks to check how many words they remember: a free play activity, a recognition task, and a word-picture matching test.

The researchers first tested the experimental model on a pilot group of 12 people; and then 50 participants joined the study. The research subjects were adult right-handed native speakers of the Russian language without any signs of neurological impairment. During the experiments, readings of the electroencephalogram (EEG) were taken. ‘Using EEG, we recorded evoked brain potentials arising in response to familiar and unfamiliar words before and after the training session,’ Olga Shcherbakova commented.

Both are good

Having analysed the data, the researchers came to the conclusion that explicit and implicit learning strategies are equally effective. ‘We have demonstrated that people successfully learn new words in either case after about ten trials. This finding was confirmed by all three verification tests,’ said the psychologist. ‘Contrary to common perceptions, empirical data, and the results of previous scientific research, explicit and implicit learning exhibit comparable effectiveness.’ And according to Olga Shcherbakova, this is definitely good news as both types of learning can be used in maintenance therapy and post-stroke rehabilitation and after traumatic brain injuries.

‘Until now, the scientific community has supported the idea that implicit memorisation occurs immediately, with no need for targeted efforts, while explicit memorising takes much effort and time. Moreover, it has been believed that in case of implicit learning a single presentation of a new word is sufficient and the learner will be able to recognise it a few months later. Consequently, there is a wide-spread opinion that any information is better recalled when it is learned in a meaningful context. Our experiment has not confirmed this theory,’ added the neurophysiologist Iurii Shtyrov, who leads the Megagrant project. Professor Shtyrov is the Head of the Laboratory of Behavioural Neurodynamics at St Petersburg University.

At the neuronal level

The situation is completely different with the topography of the neuronal mechanisms that support explicit and implicit learning. It is for this purpose that the researchers took the EEG readings. Despite the comparable effectiveness of either learning modes, the neuronal mechanisms of learning differ. During the study, the University researchers recorded different topographic distribution of brain activities. ‘With implicit learning, the left hemisphere is predominantly active, while with explicit learning – the right-hemisphere activity prevails. The implicit mode of learning appears to engage regions of the brain that have been always associated with the language. This fits well with the generally accepted idea of neuronal mechanisms,’ explained Olga Shcherbakova. ‘As for the explicit mode of learning, it appears to activate a system that is associated with metacognitive executive functions. The latter are considered a rather late evolutionary acquisition, mature and complex with regard to the amount of resources that you have to spend on them.’ The researchers from the University have been the first to be able to elucidate this issue through their experiments.

The researchers believe that the obtained data can potentially be utilised in therapy and rehabilitation of patients with brain damage. For instance, in post-stroke rehabilitation and speech therapy, the specialist will be able to choose an implicit or explicit learning strategy, depending on which areas of the brain have been affected.