CLARA: Can AI Assess Developmental Appropriateness in Children's Stories?

Sijing Yin; Zirui Wang; Qian Liu; Jiamou Liu

cs.CLOct 5, 2026

CLARA: Can AI Assess Developmental Appropriateness in Children's Stories?

Authors: Sijing Yin, Zirui Wang, Qian Liu, Jiamou Liu

Organizations: University of Auckland

Abstract

Assessing the developmental suitability of children's narratives is important for educational recommendation and developmental literacy research, yet such assessment typically relies on subjective and difficult-to-scale human judgment. This raises an important question: Can AI systems approximate human developmental judgments of children's stories? To study this problem, we introduce CLARA, a cognitively grounded framework for developmental narrative understanding through structured annotation across cognitive (COG), language (LAN), and social-emotional (SEL) dimensions, together with a bilingual benchmark resource containing 1107 Chinese--English children's stories with normalized silver developmental references and structured developmental annotations. We evaluate CLARA through benchmark comparison, component analysis, translated bilingual consistency analysis, and blinded human evaluation with educators. Experimental results show that structured developmental annotation achieves substantially stronger alignment with developmental references and human judgments than readability-based methods and direct prompting baselines. Overall, our findings suggest that AI systems can approximate certain aspects of human developmental judgment when guided by structured developmental annotation, while also highlighting the importance of interpretability and human oversight in educational NLP.

Figures & tables

Figure 1: Overview of the CLARA framework. Given bilingual story narratives, CLARA generates structured developmental annotations across COG, LAN, and SEL dimensions, aggregates developmental evidence into developmental-stage estimates, and compares the resulting representations with human and publisher developmental judgments.

Figure 2: Hierarchical developmental annotation framework across COG, LAN, and SEL dimensions.

Figure 3: Benchmark construction pipeline.

Age Group	#Mapped Stories	Share
0–2	30	2.9%
2–4	544	52.2%
4–6	687	65.9%
6–8	510	48.9%
8–12	345	33.1%
Annotated stories	1043	94.2%

Table 1: Developmental age distribution after normalization.

Figure 4: Developmental distribution of representative COG labels across age groups.

Figure 5: Developmental distribution of representative LAN labels across age groups.

Figure 6: Developmental distribution of representative SEL labels across age groups.

Model	Acc.	Cov.	L-Stab.	A-Stab.
GPT-4o	0.71 $\pm$ 0.01	17.73 $\pm$ 0.15	0.87	0.89
Gemini-2.5-Pro	0.87 $\pm$ 0.01	29.56 $\pm$ 0.65	0.66	0.79
Qwen-Max	0.73 $\pm$ 0.02	26.86 $\pm$ 0.71	0.77	0.79
DeepSeek-Chat	0.69 $\pm$ 0.02	62.41 $\pm$ 0.48	0.91	0.91

Table 2: Annotation model comparison on the 100-story validation subset.

Metric	Score
Overall Label Jaccard	0.772
COG Jaccard	0.793
LAN Jaccard	0.774
SEL Jaccard	0.775
Age Agreement	0.860
Strict Exact Full-Set Match	0.180

Table 3: Consistency under translated bilingual narratives.

Method	Overlap $\uparrow$	Mean Distance $\downarrow$
Dale–Chall	0.469	0.882
FRE	0.517	0.681
FKGL	0.603	0.502
GPT-4o	0.730	0.672
Gemini-2.5-Pro	0.768	0.497
Qwen-Max	0.775	0.467

Table 4: Benchmark comparison using normalized developmental reference labels.

Variant	Overlap $\uparrow$	Mean Distance $\downarrow$
GPT-4o Direct	0.730	0.672
Free-form + Agg.	0.791	0.411
Taxonomy Labels	0.852	0.231
Taxonomy + Mapping	0.889	0.148
Full CLARA	0.904	0.096

Table 5: Contribution decomposition analysis for CLARA.

Variant	Overlap $\uparrow$	Distance $\downarrow$
Random Mapping	0.611	0.593
Dimension-swapped	0.742	0.366
Coarse Taxonomy	0.861	0.181
Full Taxonomy	0.904	0.096

Table 6: Taxonomy quality ablation study.

Criterion	Score
COG Label Quality	4.24
LAN Label Quality	4.18
SEL Label Quality	4.31
Educational Interpretability	4.42
Predicted Age Appropriateness	4.17
Average	4.26

Table 7: Blinded human evaluation results.

Method	Human Agreement
FKGL	0.58
GPT-4o Direct	0.67
GPT-4o Rubric Direct	0.73
CLARA (ours)	0.81

Table 8: Agreement with blinded human developmental ranking judgments.

Preference	Human Evaluators
Prefer CLARA	61%
Prefer Publisher	19%
Both Reasonable	20%

Table 9: Human evaluation on disagreement cases between publisher developmental references and CLARA developmental estimates.

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Variant	Overlap $\uparrow$	Mean Distance $\downarrow$
Full CLARA	0.904	0.096
w/o COG	0.861	0.164
w/o LAN	0.873	0.151
w/o SEL	0.842	0.183

Appendix

Table 10: Dimension-level ablation study on different developmental dimensions in CLARA.

Metric	Score
COG F1	0.84
LAN F1	0.81
SEL F1	0.79
Representative Macro F1	0.73
Representative Micro F1	0.78
Human–Human Cohen’s $\kappa$	0.76

Appendix

Table 11: Gold annotation validation on a manually annotated subset.

Metric	Score
Exact Agreement	0.68
Adjacent Agreement	0.89
Cohen’s $\kappa$	0.72

Appendix

Table 12: Agreement between publisher developmental references and educator judgments.

Story Excerpt

A cloth bear receives a letter inviting him to visit the big bears when the moon is full. On the journey, the bear takes a bus, climbs a mountain path, and finally joins the celebration with the other bears.

Developmental Labels

COG: simple cause; action consequence; routine understanding LAN: basic vocabulary; story sequence SEL: emotion expression; preference expression; imitation

Publisher Reference

2–5 (mapped to 2–4 and 4–6)

CLARA Prediction

Appendix

Table 13: Representative example from the CLARA dataset.

Story Excerpt

A man is rescued by the Nine-Colored Deer after falling into a river. Although he promises to keep the deer secret, he later betrays the deer in exchange for a reward. The story follows the consequences of betrayal, trust, and moral responsibility.

Developmental Labels

COG: cause effect; prediction; sequence order LAN: narrative inference; discourse understanding SEL: emotion cause; perspective taking; rule awareness

Publisher Reference

3–8 (mapped to 2–4, 4–6, 6–8)

CLARA Prediction

Appendix

Table 14: Representative disagreement case between publisher-provided age recommendations and CLARA predictions.

0–2 years (Sensorimotor stage)
COG: visual tracking, sound response, face recognition, object recognition, object permanence, color recognition, shape recognition, size discrimination, pattern recognition, sensory integration
LAN: sound association, babbling pattern, word recognition, word-object mapping, phonetic awareness
SEL: emotion reaction, emotion recognition, attachment security, social smiling, caregiver response
2–4 years (Early preoperational stage)
COG: simple cause, action consequence, routine understanding, object function, basic counting, category learning, similarity detection, difference detection
LAN: basic vocabulary, two-word sentence, sentence imitation, naming objects, story sequence, picture description, symbolic play language

Appendix

Table 15: Full developmental label taxonomy used in CLARA.

Explore similar work

Evaluating Developmental Cognition Capabilities of LLMs
May 8, 2026Xiao Xiao, Hayoun Noh, Mar Gonzalez-FrancoConversational Artificial Intelligence Developmental Trajectories
De Vinci Research Center, France · University of Oxford, UK · Google, USA
BIASEDTALES-ML: A Multilingual Dataset for Analyzing Narrative Attribute Distributions in LLM-Generated Stories
Apr 18, 2026Yuxuan Ouyang, yingfeng luo, JingBo Zhu +1Narrative Generation Narratives
School of Computer Science and Engineering, Northeastern University, Shenyang, China · NiuTrans Research, Shenyang, China
CraftAlign: Feature-Grounded Evaluation and Revision Guidance for AI Stories
Aug 2, 2026Yang Yang, Boyun Xu, Shaofeng Liang +5Value Alignment Multi-Dimensional Evaluation
The Hong Kong University of Science and Technology (Guangzhou), Guangzhou 511400, China · Institute of Deep Perception Technology, JITRI, Wuxi 214000, China

May 8, 2026, cs.AIJ/K move · Enter open · S save

Evaluating Developmental Cognition Capabilities of LLMs

Xiao Xiao, Hayoun Noh, Mar Gonzalez-Franco

De Vinci Research Center, France · University of Oxford, UK · Google, USA

Conversational Artificial Intelligence Developmental Trajectories

Open paper AskPDF arXiv

Conversational AI is increasingly personalized around users' preferences, histories, goals, and knowledge, but much less around how users interpret and take up model outputs to construct and understand their reality. We draw on Robert Kegan's constructive-developmental theory as a complementary lens on this dimension. Existing methods for assessing developmental stage in the Keganian tradition rely either on expert interviews that do not scale or on sentence-completion instruments that are proprietary, lengthy, or invasive. To make this perspective tractable for LLM evaluation, we introduce the Developmental Sentence Completion Test (DSCT), a 20-item instrument designed to elicit developmental signal in self-administered text. Throughout, we treat the resulting labels as characterizations of stage-like structure in elicited responses, not as validated person-level developmental stage. We then ask how much of that signal can be recovered by LLMs across three elicited response regimes: simulated personas, real human respondents, and default model-generated answers. On simulated personas, top frontier models recover simulator-intended labels with high accuracy. On real human DSCT responses, human-LLM agreement is fair, with much stronger within-neighborhood than exact agreement. Finally, when LLMs answer DSCT prompts without persona-conditioning, their responses exhibit stable stage-like differences across model families, with larger and newer models tending to generate higher-rated text. These results suggest that stage-conditioned signal is cleaner in synthetic responses than in human-written DSCT text, and that the core constraint for stage-aware conversational AI is not classifier accuracy alone, but the availability of developmental signal from elicited text.

Abstract

Explore similar work

Evaluating Developmental Cognition Capabilities of LLMs

BIASEDTALES-ML: A Multilingual Dataset for Analyzing Narrative Attribute Distributions in LLM-Generated Stories

CraftAlign: Feature-Grounded Evaluation and Revision Guidance for AI Stories