Humboldt-Universität zu Berlin - CALLIDUS

Concept of the vocabulary unit

As a special feature of the software (Machina Callida) there is a prefabricated vocabulary unit, which can be used to inspire a variety of vocabulary work in Latin lessons.

Some hypotheses on vocabulary acquisition

  • Words also derive their meaning from context ("polysemy").
  • A text is opened up a) via form (similarity to already known words, bottom-up), b) via the existing text-internal and text-text-internal background knowledge (top-down) and c) via the generation of sentence structures that match these references (without using the sentence structure for control). Therefore, all sub-strategies need to be practiced, with step c) being used primarily to review the conclusion.
  • Vocabulary competence consists of three components:
    1. declarative knowledge (form, function (meaning, application), strategy): What does the word x mean? (knowledge)
    2. procedural knowledge (contextual retrieval, learning and development strategies): How can I determine the meaning of word x in context? (Automation)
    3. executive knowledge (self-control, reflection, ...): Have I observed all the necessary rules for determining the meaning of a word in context? (Metacognition)
  • The more multifaceted a word is learned, the more lasting its networking is in the mental lexicon and the faster and more reliable the retrieval of a meaning is.
  • Exercise types should preferably fit the type of vocabulary acquisition: If the focus of language acquisition is on reception (recognition), the exercise and test formats must reflect this.


Objectives of the unit or research questions

  • Illustration of how the software (Machina Callida) could be integrated into the vocabulary work in the school.
  • Test mode --> Data collection to answer the following questions:
    1. What requirements must be met for learners to master a Latin cloze?
    2. How important is text comprehension in order to be competent with the exercise format "Latin gap text"?
    3. Are there differences in success in the vocabulary test depending on the exercise format (gap text or word list)?
    4. Is there a connection between text comprehension and success in the vocabulary test?
    5. Which query formats do learners master well, which less well?
    6. How much time do learners need on average for the individual sections? Are they reasonably extensive?
  • Learning mode (see also test mode):
    1. How much time do learners need on average to complete the individual tasks (without timer)?
    2. Is there a connection between slow processing and the learning outcome?
    3. When is a task canceled?


Structure of the vocabulary unit

Text selection: Cicero, letter to his brother Quintus (governor in the province Asia)

Justification: Cicero is a central author in the upper school, learning vocabulary and school grammar are oriented to him. At least in Berlin-Brandenburg, the collection of letters is also relevant to the Abitur (Q1); the selected letter to Quintus allows an insight into the relationship of the brothers to each other and into the political work of the two Ciceros.

Vocabulary selection, general: intersection of the Cicero excerpt and basic vocabulary (Bamberger Wortschatz = BWS)

Vocabulary selection, especially: The words or word groups that are particularly important in the context or occur more frequently (even with different meanings) were selected for the tasks (pronouns).

Sequence of the individual elements:

  1. Pretest (test mode only): 5 different task formats on different aspects of vocabulary knowledge
  2. Introduction to the text: brief introduction, marking of verbs on the Latin text --> first contact with the original, methodical preparation of the text opening (similar to transphrastics or linear decoding)
  3.  Text comprehension: tasks for text comprehension: bilingual text (Latin text documents), monolingual English text for overall comprehension (multiple choice)
  4. Enter gap text or word meanings for exercise types: Words or word groups of the preceding tasks are repeated here, once in the Latin context (gap text), once via form meaning associations (input of a word meaning).
  5. Posttest: identical to pretest
  6. Evaluation: entrance test (test mode only), vocabulary work on the text, exercise, final test (+ relative change to entrance test); only the actually completed tasks are taken into account in the evaluation.

Justification: The unit simulates teaching events on a small scale. At the beginning of a unit a vocabulary test can be used to test the vocabulary knowledge relevant for the reading (diagnosis). The unit then begins with an introduction to the unknown original text (Cicero letter) and a linguistic task to predefine the text (marking predicates). In this way, the Latin text is perceived for the first time. The marking of the predicates should already put the focus on the conscious perception of the words, even if (depending on the learning group) only little is understood altogether. In the next (exercise) phase, the learners deal more closely with the original text by finding and indicating Latin text documents for certain formulations or statements based on the German translation. In order to provide sufficient clarity (e.g. also on smartphones) and above all to ensure a (cognitively) well manageable scope of tasks, the text was divided into six sections. This, however, gives rise to the fear that the overall meaning of the text excerpt will be lost. For this reason, learners should mark the correct statements in a further text comprehension task (multiple choice), whereby they receive the translation of the entire text as assistance. The exercise phase then begins (word selection: words focused on in text comprehension) either with an intralingual gap text or with a (randomly generated) word list. The exercises are selected at random. These exercise formats reflect two typical learning phases in Latin teaching: the development of word meanings in context (during translation) and (repetitive) learning according to a list (e.g. for the preparation of tests). Finally, a check is made of the words or groups of words that have been thematised several times. Since none of the learners may be preferred by their given form of exercise (test theory), the words are checked neither in the test format of the pair associations (Latin - English) nor of the gap text, but with other test formats which are oriented towards different partial competences of the vocabulary competence.


Exercise formats

As a rule, these are closed task formats that only allow one possible answer (occasionally with several variants of meaning). The correct spelling must always be observed, only the upper and lower case does not matter. The progression of the exercise formats can be recognized by their arrangement. For technical reasons open exercise formats are missing, which would allow a free answer, e.g. a justification of the selection, since an automatic evaluation of the results takes place.

The exercise formats are presented separately.




The course should take place in a 45-minute lesson. The use of smartphones, tablets, laptops and smartboards is explicitly supported, but work on the PC is also possible. Particularly helpful for the improvement of the software would be the group-wise run through the test mode.


The evaluation takes place anonymously. During the practice an ID for the individual participant is automatically generated in the background, which he keeps as long as he follows the vocabulary unit, so that his individual result can be evaluated. This ID then expires. It is linked exclusively to the process of "learning" and not to traceable personal information (not even to the IP address of the Internet access). This ensures the highest possible level of data protection. However, every abort (x-button top right) of a task leads to the deletion of the data record, i.e. you cannot interrupt the work and resume it later at this point. Even sending the exercise results does not happen automatically, but must be triggered by clicking the Send button at the end of the unit (bottom left). In this way the right to informational self-determination of the learner is preserved.