![Sample our Behavioral Sciences journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/110801/TOC-Subject-Sample-2021-Behavioral-Sciences.jpg"})
ABSTRACT
Background: Aphasia patients can produce some words reliably but have difficulty retrieving other words. In this article we review one of the variables that have been proposed to explain which words remain available.
Aims: Age of acquisition (AoA) refers to the age at which people learn a word. We review the evidence indicating that early-learned words remain more accessible than late-learned words in acquired aphasia. We examine the most likely mechanisms.
Methods and procedures: We review the various studies that investigated the effect of AoA on word retrieval in acquired aphasia and in dementia. We link the findings to differences in processing efficiency in healthy individuals and to the mechanisms that have been proposed there to explain the differences. We also have a critical look at the methods used in various studies and investigate which precautions must be taken to advance our knowledge about the AoA effect.
Outcomes and results: We argue that the effects of AoA and frequency are a result of the brain acquiring information over time in an incremental manner, improving with practice and possibly showing a decline in plasticity as the brain ages and more information becomes stored. Because AoA is yoked to frequency, an AoA effect is likely to be accompanied by a frequency effect of roughly the same size. In addition, AoA has a unique effect on the ease with which the correct verbal response can be retrieved in a naming task based on semantic input, such as object naming.
Conclusions: Words learned early in life are more likely to be retained in acquired aphasia than words learned later in life. The finding that AoA has a unique effect on the ease with which the correct verbal response can be retrieved in object naming may be due to the organisation of the semantic system, such that early acquired meanings are richer, more accessible and more robust against brain damage. Alternatively, it may be due to the way in which semantic information is translated in verbal output. To further our knowledge, researchers are encouraged to use more powerful regression designs with more stimuli, and to include better measures than has been done in the past. We also argue that more information can be gathered by not limiting the research to picture naming.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Studies in which this was not the case often used sub-optimal measures of frequency or AoA, or did not have enough variation in one of the variables (see later).
2. In all the analyses we did, subtitle-based frequencies also outperform frequencies based on spoken corpora (Brysbaert & New Citation2009; New, Brysbaert, Veronis, & Pallier, Citation2007; see also Ernestus & Cutler Citation2015, who compared subtitle frequencies with spoken frequencies for auditory lexical decision). The reason for this is that spoken corpora tend to be rather small and restricted in the variety of contexts sampled and topics talked about.
3. Among whom one of our reviewers.
4. The authors thank Philip S. Dale for kindly sharing the data with them.
5. This word frequency measure based on child directed speech was used by Goodman et al. (Citation2008) and is more representative for the world of young children than frequencies based on adult directed materials.
6. So, the Morrison et al. (Citation1997) objective AoA estimates are becoming increasingly interesting for studies with older participants and will remain so for at least another 50 years.
7. An example of a true experiment with words is semantic priming. In such studies the same target word (e.g., “doctor”) can be assigned at liberty to the semantic priming condition (preceded by the prime “nurse”) or to the control condition (preceded by the prime “purse”). Finding a difference between both conditions in a properly designed study then allows the researchers to accept the effect, independent of the target words used.
8. Another variable that may matter and that has not yet been investigated is the number of morphologically derived words and compound words the name of an object has, a variable known as the morphologically family size (Baayen, Feldman, & Schreuder, Citation2006). Words with a large family size are easier to process than words with a small size.
9. In this context, we note that Bastiaanse, Wieling, and Wolthuis (Citationin press) reported an effect of word frequency on retrieval of nouns but not verbs with AoA and imageability controlled.