Hippocampus activity differentiates good from poor learners of a novel lexicon

Abstract

Language proficiency is a key to academic and workplace success for native and non-native speakers. It is largely unknown, however, why some people pick up languages more easily than others. We used event-related functional magnetic resonance imaging (e-fMRI) to elucidate which brain regions are modulated during the acquisition of a novel lexicon and which of these learning-related activity changes correlated with general semantic language knowledge. Fourteen healthy young subjects learned a novel vocabulary of 45 concrete nouns via an associative learning principle over the course of five blocks during e-fMRI. As a control condition, subjects took part in a structurally identical “No-Learning” condition lacking any learning principle. Overall, increasing vocabulary proficiency was associated with (intercorrelated) modulations of activity within the left hippocampus and the left fusiform gyrus, regions involved in the binding and integration of multimodal stimuli, and with an increasing activation of the left inferior parietal cortex, the presumed neural store of phonological associations. None of these activity changes were observed during the control condition. Furthermore, subjects who showed less suppression of hippocampal activity over learning blocks scored higher on semantic knowledge in their native language and learned the novel vocabulary more efficiently. Our findings indicate that (a) the successful acquisition of a new lexicon depends on correlated amplitude changes between the left hippocampus and neocortical regions and (b) learning-related hippocampus activity is a stable marker of individual differences in the ability to acquire and master vocabularies.

Introduction

Language proficiency predicts academic and socioeconomic success in both native and non-native speakers (Greenberg et al., 2001, Kaestle et al., 2001, Young et al., 2002). With increasing transnational migration of workers in most parts of the world, achieving high levels of proficiency in a foreign language becomes economically, socially, and emotionally pivotal for an increasing number of people. The biological and cognitive mechanisms contributing to individual differences in learning and mastering languages are beginning to be unraveled (Adams and Gathercole, 2000, Baddeley et al., 1998, Callan et al., 2003, Cornelissen et al., 2004, Musso et al., 2003, Opitz and Friederici, 2003, Perani et al., 2003, Raboyeau et al., 2004, Wang et al., 2003), but it remains elusive why some people rapidly evolve into language experts while others remain novices. The search could be advanced by identifying brain regions whose activity predicts differences in language learning success.

Language comprises both a mental lexicon (semantics) and a set of grammatical rules to combine the words into a sentence. Here we concentrated on the mechanisms underlying the acquisition of novel vocabularies because learning of a lexicon constitutes the first essential building block of language learning in both children and adults (Pitchford et al., 1997). Behavioral observations of normal children and adults (Markson and Bloom, 1997) and lesion studies (Ullman, 2001) suggest that the ability to learn a novel lexicon is mediated by neural networks that extend beyond classical language regions. However, functional imaging studies to date have not directly investigated the neural processes involved when novel words acquire meaning. Two recent studies, one using positron-emission-tomography (Raboyeau et al., 2004) and one using magnetoencephalography (Cornelissen et al., 2004), compared brain activations prior to and after several weeks of training in naming concrete objects in a novel language. Because the respective activation tasks assessed retrieval of the newly acquired words, the design of these studies does not provide answers to the question which brain structures are involved in the initial acquisition of a novel lexicon.

Declarative memory can be episodic (time- and place-specific experiences) or semantic (facts and general word knowledge). A critical role of the hippocampus proper for the encoding of episodic memories is supported by both animal and human studies (Squire and Zola Morgan, 1991, Squire et al., 1993). The contribution of the hippocampus to storing semantic memories like the mental lexicon, however, has been a matter of debate (Squire et al., 2004, Tulving and Markowitsch, 1998). Human lesion studies support the view that the acquisition of semantic information is initially not separable from an episodic event of which it is part, and that encoding into either memory system initially depends on the hippocampus (Manns et al., 2003, Squire and Zola, 1998), at least for achieving normal levels of performance (Holdstock et al., 2002). This view is additionally supported by recent functional imaging studies with healthy subjects: irrespective of whether information is episodic or semantic, the hippocampus seems to be particularly important in the initial binding of new multimodal associations in a variety of domains, like learning to associate a certain face with a name (Bernard et al., 2004, Brasted et al., 2003, Davachi, 2004, Sweatt, 2004) or the initial incidental acquisition of factual information (Maguire and Frith, 2004).

Associative learning contributes to the acquisition of a novel vocabulary in children and adults (for an overview, see Breitenstein and Knecht, 2003). Work by Saffran (2001) and Saffran et al., 1996, Saffran et al., 2001 shows that both children, starting at the age of 8 months, and healthy adults can be trained in identifying words in a continuous sound stream purely by using associative principles implemented in this sound stream (Saffran et al., 1997). It has furthermore been demonstrated that 14-month-old children learn novel object names the better the more frequently a particular object name is mentioned in the presence of a respective object (Gershkoff-Stowe, 2002). Finally, there are several studies on the implicit acquisition of grammatical rules based on associative characteristics embedded in the stimulus material (Braine et al., 1990, Gomez and Gerken, 1999, Gomez and Gerken, 2000, Knowlton and Squire, 1994, Knowlton and Squire, 1996, Marcus et al., 1999, Morgan et al., 1989, Reber, 1989, Saffran, 2002, Tunney and Altmann, 2001). These findings demonstrate that the general cognitive mechanism ‘associative learning’ is essential for language acquisition. Given the importance of the hippocampus in this kind of learning, we asked specifically whether the hippocampus also contributes to the acquisition of a novel vocabulary.

In summary, we aimed to elucidate the neural network mediating the acquisition of a novel vocabulary. Of particular interest was the role of the hippocampus during the initial acquisition of a novel vocabulary as part of the semantic memory system. We additionally asked whether learning-related activation strength in any of the contributing brain systems predicts how well subjects acquire a novel language and how well they master semantic language aspects in everyday life. The latter aspect was examined by correlating individual brain activations with tests of general verbal semantic abilities which were assessed outside the scanner.