Learning & Teaching Foreign Languages

U-shaped Learning

Read and Reflect

The so-called U-shaped learning curve involves an initial targetlike production, followed by a dip in performance, which then recovers.

  1. U-shaped learning (Case)
  2. developmental versus instructional sequences (Lightbown)
  3. transfer effects (Kellerman)



John Case, a researcher in Computer and Information Science at the University of Delaware, provides a brief sketch of the u-shaped curve, common in first language acquisition, and also found in second language learners.

The unusual phenomenon of U-shaped learning is found in early childhood cognitive development. For example, a child begins to use the verb ‘spoke’, subsequently, she may use instead ‘speaked’, and, later and finally, she again uses ‘spoke’. Her learning follows a U- shape as in the figure below.
Graph showing U-shaped learning curve


A Canadian second language researcher who studies classroom SLA, Lightbown showed that French learners of English lost accuracy in progressive -ing forms as they added the simple present/present continuous distinction to their tense/aspect systems. 

In early stages, learners used the progressive marker for both simple and continuous aspects (overgeneralisation). 

As they learned not to mark simple present verbs, progressive marking in continuous contexts also dropped, leading to more errors where previously there were few. 

In the third stage of acquisition, learners once more used the -ing form to mark the present continuous correctly, but not in contexts requiring the simple present.


Graph showing aquisition of -ing progressive

Diagram from Susan Guion Anderson at the University of Oregon http://darkwing.uoregon.edu/~guion/440notes/ellis6.html


Another example of U-shaped learning is provided by Kellerman's 1985 research on Dutch learners of English.  In his experiment, Kellerman asked the Dutch learners to decide whether sentences like "he broke the vase" (a transitive construction) and "The vase broke" (an ergative construction) were acceptable in English.  Both constructions are possible in Dutch.

1. Can you imagine a U-shaped learning curve for the acquisition of the verb "broke" in transitive and ergative contexts?

2. What might cause the dip in performance, or backsliding?


1. Can you imagine a U-shaped learning curve for the acquisition of the verb "broke" in transitive and ergative contexts? 

Kellerman's study showed that younger learners accepted both uses of the verb: they judged "He broke the vase" and "The vase broke" to be good English.  Older subjects, aged 17-18, rejected the ergative construction: they judged "The vase broke" to be incorrect English. Kellerman's oldest learners, aged 20 and above, accepted both transitive and ergative uses.

2. What might cause the dip in performance, or backsliding?

As learners integrate more sophisticated linguistic competence, they may reject previously accepted forms as part of the process of restructuring their interlanguage. The marked status of ergativity might act as a brake on use of L1-like constructions: "The vase broke" is a less common construction than "He broke the vase." Further input and experience may lead the learners to accept the less common, L1-like ergative forms once again.

Valid XHTML 1.0 Transitional